Monday, March 31, 2008

Google Docs Going Offline

I'm really late to this party. Both the Official Google Blog and the Official Google Docs Blog announced offline support coming in Google Docs this afternoon, and this is the first chance I've gotten to write about it. So I'll publish a brief description (more details are of course available from Googlified, Google Operating System, and Lifehacker; strange lack of Google Blogoscoped so far) and a short commentary.

Everyone has probably heard of Google Gears by now, that open-source browser extension that lets Web applications enable offline functionality. Google Reader uses it (until now the only Google service to do so), as does Remember The Milk (which I love, except for the flawed sharing). Supposedly Google Calendar is in the works, and Haochi said Gmail is "possibly" going to get it.

Google Docs' offline functionality works only for documents at the moment; spreadsheets and presentations will have to wait. And it's only available in English. Rather than rehashing everyone's prose, I'll just shamelessly use the same YouTube video. (It is featured on nearly all the sites I mentioned above, BTW. I wish Google had continued using Google Video embeds though.)



So now you know how it's supposed to work. I can't test it myself because I haven't got it yet -- and I probably won't for a while, judging from past experience with Gmail IMAP. Update (04/01): This paragraph should have mentioned a few restrictions. The feature is only available in English for now, and it's being rolled out to everyone over the next few weeks, which means most people probably won't be able to use it for a while, like me.

What I'm glad about is that it -- at least purportedly -- continuously saves the data required to go offline, like RTM does. There's no planning to go offline like Google Reader makes you go through. (It always annoyed me how you couldn't get images, even, offline; but I digress.)

To summarize, it looks like the offline feature will be killer. I know Zoho Writer has had this for a while, including the ability to edit offline, but they started with just viewing; Google took longer and launched with all the functionality they could. And Zoho's suite, as far as I know, still requires you to manually prepare to disconnect from the tubes.

This battle of the online office suites gets more interesting by the day. I wonder what's in store next week...

Update (04/19): Some time in the last week or so, I got access to this feature. I don't use Google Docs every day, so I'm not entirely sure when it showed up. But it wasn't there at the beginning of the week; I'm sure of that. Anyway, it doesn't seem to be especially useful, really, because of one little caveat: No new documents. Harrumph. Fortunately I'm not offline much.

User Actions Shouldn't Interrupt Critical Stuff

Yes, I know that's a rather cryptic title. I'll do my best to illustrate what I mean in the coming prose. This was all inspired by something that happened to me last night while using the Blackboard Learning System software (it's Web-based) to try and take an assessment in one of my courses for school. (Inspiration from homework? It's actually happened before, like when I blogged about annoying network-storage habits.)

All right, down to business. Specifically, what happened last night was that I tried to perform an action on the page before it had finished loading. The tests in Blackboard (at least recent versions of the software) have a bar near the top that contains status information on each of the questions, but it is collapsed by default. Clicking the header makes it open up and actually display useful information. Simple JavaScript, but the action of expanding that box had catastrophic results.

Oh, it expanded just fine; there's no question that it functioned correctly. But that function wasn't isolated. Immediately upon my click, the entire page (well, frame; Blackboard still uses that anachronistic design feature) ceased loading. Now, that was a problem, because the buttons to manage the test are at the very end. Uh-oh.

I had before me 25 out of 29 questions, with no way to save my answers and continue, nor to submit the test incomplete. My curiosity definitely caused a problem. But it didn't have to be like that.

Unfortunately, I don't have the luxury of scrutinizing the source code of the page for this post. That would necessitate sacrificing another test attempt (and would require me to send another email to my teacher requesting a reset). I can still theorize as to why the problem happened, and brainstorm possible solutions.

First, the cause. While I can't access the page source, as I said, I suspect there was something in the way the expansion link was constructed that caused a navigation event in Internet Explorer (Blackboard doesn't always play nice with Firefox, in my experience, so I use IE6).

Many JavaScript effects are created using an onclick attribute and href="#". I know that when I click a hyperlink on a page in Firefox, anything still loading on the current page is dropped and the new site begins to load. Internet Explorer probably does the same thing. By clicking the link to expand the box, I may have caused IE to try and find an unnamed fragment on the page; and since even navigating to a fragment counts as a navigation, I probably stopped the page loading myself.

I'm not saying that's what had to happen -- that there is no way to prevent it from happening. I'm merely theorizing. If the function or statement called by onclick returns false, the browser shouldn't navigate. That's what Firefox does, at any rate. So perhaps there was a missing or incorrect return statement in the function call. It's also possible to use another tag, like a <span>, to create the link and style it with CSS. The onclick is valid on every element I can think of, and the superfluous href wouldn't be a problem.

Aside from changing the way that box is coded, there is another easy solution to the problem: The "Save" and "Submit" buttons could be duplicated at the top of the assessment page. This is certainly not the first time I've had this problem; it's just the first time it's happened to me on the initial load of a test. (I can usually go back to a previous page and try again to load the refreshed version of the test.)

I believe this might be considered a race condition, in that the behavior of the assessment page is dependent upon when user interaction begins. Race conditions are nasty things, of course, and they've become the bane of any programmer's existence. I'm new to the concept, though, since I've been fortunate enough to avoid them in my own work. I usually hear about them in the context of JavaScript, come to think of it.

The take-away message from this post is not that Blackboard itself has a big problem. I've gone quite a while without this being an issue (nearly four years, actually). The real issue is that this could happen in a more critical situation, like a Web-based management console for, say, a nuclear reactor or weapons system. (More realistically, flight-control or remote monitoring, but the others sound more dire. ;-) An error such as this could cause enough of a lag that an operator would be unable to correct a problem in time and the system could cause a disaster.

In short, the moral of this story is that user interaction shouldn't interfere with critical internal functions in an application, whether Web-based or otherwise. The next time something like this happens, the consequences could be more than a simple locked test.

If you're interested in a story created around a similar event, be sure to check out the fictionalized version of events on The Queiba Wars. CodingExperiments.com has a more general summary of this kind of problem as well, should you be interested in that.

Sunday, March 30, 2008

FriendFeed Improves in Leaps and Bounds

Since going public last month, FriendFeed has made numerous advances in functionality and service support. In March alone, there have been six posts on the FriendFeed blog, some containing news of multiple enhancements and/or additions to the service. Here's a run-down of all the improvements FriendFeed's made in just one month.

The first thing to happen was a speeding-up of the service. Bret Taylor posted "FriendFeed is now a lot faster" on March 11, and indeed the site was blazingly fast from then on. It still is. The joke at the time was that a new Ethernet cable sped it up (FriendFeed thread), but the truth is that there was a lot of code tuning, caching, and a new indexing scheme added (FriendFeed thread, changelog entry).

Three days later, feeds for comments and Likes were added. March 14 was the day you could finally "See the stuff your friends commented on and liked". There were also some nice statistics added: Comment and Like totals for both the past week and all time became available in the bar on the right of every user's profile page.

On March 17 (yes, another three-day interval), FriendFeed got search! Not only could you now see your friends' comments and Likes, and not only was the site now blazingly fast, but items previously lost to the ether could now be found again. (I call the addition of search the "Googlification" of FriendFeed, because nearly every Google service has a search box.)

Between the addition of search and the next update, there was a one-week reprieve. It was good to have a rest from FriendFeed update stories because it made the next announcements all the more wonderful. March 24 saw a boatload of updates. First of all, there were five new services added (Disqus, Goodreads, LibraryThing, Seesmic, and SlideShare). Second of all, there were numerous bug fixes. Third of all, FriendFeed got the ability to post back to Twitter! Commenting on a tweet in your feed gave you the option to also send the comment as an @reply via Twitter.

As if all that wasn't enough, the FriendFeed team launched their API the following day, on March 25. Since the launch, third-party developers have released a WordPress plugin (which can be seen in action on CodingExperiments.com) and a statistics site, with an Adobe AIR application rumored to be on its way.

So, in short, FriendFeed is making humongous strides in advancing its service. They sure turn around features a lot faster than Google... Here's to another month of FriendFeed advancement!

Oh yeah, it can't hurt to plug my FriendFeed page while I'm talking about the site, can it? ;-)

Saturday, March 29, 2008

Move to Gmail from an Offline Client?

Various methods for switching to Gmail and bringing along your old mail have been discussed to death in the blogosphere, I know. Gmail does turn four next week, after all (it was released to the public on April 1, 2004; April 1, 2008, is next week). Since its release, Gmail has added lots of features, including IMAP support.

Ordinarily I wouldn't bring up something that's already been done before (well, anything that was last posted about more than a week ago). However, the Official Gmail Blog posted a short set of tips for moving to Gmail, and I thought I'd sound off on the list. (Comments are turned off on that blog, and I need a post anyway. ;-)

So, despite the title ("Tips for importing old email to Gmail"), the only things we find out are that we should set up POP retrieval on all our old accounts and enable sending messages as those old addresses just in case. Nothing is mentioned about getting mail from email clients, which has been deleted from the POP server, into Gmail. And nothing is mentioned about getting mail out of e.g. Hotmail or Yahoo! Mail (which are, admittedly, pretty difficult as they don't support POP access anyway).

Various solutions for the desktop-client problem have been posted in the years since Gmail's launch, as I hinted earlier in this post. (I haven't seen any good solutions for Hotmail/Yahoo! Mail yet.) Some rely on automatic forwarding scripts, others on the new IMAP support (Zoli Erdos has a good write-up of the IMAP method, which I would like to try when I get the chance).

What irks me is the fact that none of these ingenious solutions for getting already downloaded mail into Gmail are mentioned in the Gmail Blog post. If you use a client like Outlook Express or Thunderbird (Mail.app, Outlook, Eudora... the list goes on), Google can import your stuff if you get a business-class Google Apps account, which supports email migration. But that would be $50 a year. (The migration API on which the migration tools are based is only available on Google Apps-based Gmail accounts, unfortunately.) Granted, the solutions that simply forward everything to your account and ruin all the headers probably aren't all that useful; but the IMAP-based methods would probably be quite useful. It's just a drag-and-drop from local mailbox to Gmail. Why didn't the post mention them?

I hope that tips for uploading email from offline clients get posted to the Gmail blog sometime in the future. Preferably in the next few months, but anytime in the next up to two years would be fine by me. It had better be before Gmail's tenth birthday, though.

Oh, and rest assured, I will try the IMAP import method as soon as I have the time. That will probably be around June, after school's done for the year. Cross your fingers for me in advance? (Thanks if you did! :-)

Update (04/01): Just got a comment from Zoli Erdos, who brought my attention to the text surrounding the link to his site. I cleaned that up; it didn't make much sense before. So this update is really just a short series of edits. I've marked my inline insertions up with <ins> tags for visibility. Thanks for bringing me back here, Zoli! :-)

Friday, March 28, 2008

Transfer Google Document Ownership

I'm not talking about handing over copyright certificates to publications produced by the search giant. It would be cool if us little people, Google's loyal users, could get a piece of their intellectual property, but no. I'm talking about making someone else the owner of a Google Document you created.

Google Docs has had sharing built into it since I started using it many moons ago. However, the owner of the document has always remained static until now. I just noticed late yesterday (read: too late to blog about it) that the "Change Owner" option Ionut discovered (look below the screenshots) a couple days ago. I didn't see it then, but I see it now!

So, where is this mysterious option? I had to actually do some digging to find it. It's not in the sharing menu, or any of the other menus available when the document is open (that I could see). It's only available in the context (right-click, but you knew that ;-) and More Actions menus on the main Docs screen:





It's at the end of both menus (which are copies of each other, so consistency is expected). Kind of hidden, which is probably OK because not a huge number of people will have need of the feature. I do see how it could be useful, say, if you start a document and then delegate it to someone else, and further collaboration from you is unlikely. No sense having it still in your list. The other person can take over responsibility for the Doc. Transfer it.

Niche feature, niche product; but that's OK. Microsoft Office has been packing niche features, dozens of them, for years, and the numbers keep growing. Now if only Google would implement that nice paged layout we saw on Google Blogoscoped a little while back (scroll down to "A Newer Google Docs" - the post is about a whole presentation Google made)...

Thursday, March 27, 2008

McAfee Destroyed SiteAdvisor

You may or may not have heard of SiteAdvisor before. I've mentioned it a few times on this blog, but usually in passing. Say, when comparing it to WOT. This post focuses on McAfee SiteAdvisor, and the reasons I've dropped it.

First off, it used to be an independent company. Then it was acquired by McAfee, a company that makes anti-virus software (subscription-based). Before the acquisition, everything was free. After the acquisition, McAfee removed some features and put them into a paid version of the program available as part of their Internet security suite.

But that wasn't enough. They also have to nag users of the free version once a month, the first time they visit the website for more detailed information. A splash page is displayed before the website rating, asking if the user is sure they don't want to upgrade to the paid version of SiteAdvisor. Once every 30 days.

And another thing: I don't remember ads on the original SiteAdvisor site. In fact, much of the layout has changed since the McAfee acquisition (even the favicon, which is now the McAfee M; I don't even remember what it used to be). Now there is a blinking, animated ad at the top of the page. I don't see it usually, since I run AdBlock Plus (fabulous Firefox extension, BTW), but I know most users do. Why not use Google AdSense for unobtrusive text ads? Don't they pay enough?

In short, the user experience has degraded quite a bit, at least from my perspective. A related concern is an incompatibility between the SiteAdvisor extension and the (great but currently non-functioning in Gmail because of code changes) Freenigma extension which caused a non-functioning address bar for me. Both add-ons are disabled now, as I've stopped using both in everyday surfing, but it's still an incompatibility. The last update was months and months ago, long before even Google Talk's last revision.

Obviously, I've replaced McAfee SiteAdvisor with WOT. I just hope WOT doesn't go the same direction as SiteAdvisor, perhaps after an acquisition by Norton (so they can compete with McAfee, of course).

Wednesday, March 26, 2008

Google Docs Gets Colored Folders

Apparently I was a little too hasty to publish my announcement of the new menus. I didn't wait long enough before posting to find another new Google Docs feature, colored folders.

Gmail's had colored labels for a while now (at least in the new version). Google Docs has been going through iterations left and right lately; one wishes some of that time would be spent on Google Talk...

Anyway, so here's an example of the colored labels folders (I really wish they'd just make them labels again):



The folders display in the list just like Gmail's colored labels. (In fact, I'll bet they used the same code.) I did post a screenshot of the Gmail interface with coloring back when that feature was released. It might be different, though, only using snippets like the color picker. Gmail doesn't freeze my browser when changing labels, and Google Docs does temporarily lock up Firefox for me when I change folders now (it didn't before). I wonder what's up with that...

Now, of course, the question is: Will Google launch another new thing today and keep me busy blogging, or will I have a chance to ignore the Internet for a bit and catch up on my reading?

Google Docs Gets Menus

Apparently commands in Google Docs weren't organized well enough. Actually, it was more than apparent. Things that would take me milliseconds in Microsoft Word (on account of the fact that I've used it so long) used to take seconds in Docs. The menu structure just wasn't intuitive.

Well, the Docs team did some card-sort studies (part of usability research) and figured out a better way to organize menu commands. Instead of just having one File menu, the top of the application is now graced by a full complement of menus, Office-style (Microsoft pre-2007 or OpenOffice.org, take your pick; they're pretty similar).

I'd say it makes for a pretty neat interface (even better than the last redesign). And what's more, they've brought back the old Save/Save & Close buttons. Sadly, the Discard Changes button is still MIA. I liked that; it kept me from saving unnecessary revisions. Oh well.

So, behold the new toolbar, with menus:



Slick, eh? Now all they have to do is port the changes to Spreadsheets and Presentations. :D

Tuesday, March 25, 2008

Financial Reports Waste Paper; Use the Internet

Financial reports are a colossal waste of paper. Well, perhaps that's not entirely true. I'm sure some people read every last character in those thick stacks of paper sent out by investment agencies at the end of the fiscal year. But most of us just shred them and put them in the recycling (or, unfortunately and probably more commonly, the trash).

What's in one of those things? Usually it's just a summary of all transactions conducted by the agency on your behalf since the last report, with a few statistics thrown in for good measure. Yes, it can be useful. No, it does not need to be mailed out to every agency customer, in triplicate (I kid you not; this does happen on occasion).

Before the invention of the Internet, such mailings made sense. There was simply no other good way of getting the documents. It was either paper mail or paper fax. Either way, you used paper.

Now, of course, we have the Internet, and that wonderful thing called email. Adobe made the PDF file format, and we should be using that, too. There's simply no good reason to use paper for these reports in this day and age.

Considering that each report averages around 20 or 30 pages, and that there will often be multiple reports mailed by multiple agencies, that translates into a lot of paper, especially if the amount is multiplied by the number of customers receiving those documents. As I said, we could definitely email PDFs instead of putting all these reams of paper in the mail. What if, just hypothetically, there was a better way? A better way even than email?

I think there is, and it's kind of making me eat my own words. What if all the reports were simply generated on-the-fly by a Web-based console? Each customer would get a login (always transmitted over SSL, of course, as with the rest of the site), which would get them into a reporting interface tied to their transactions. Undoubtedly, the transaction data is stored in a database. That would be an efficient way to store such uniform data. The site would simply tie into that database (or a replicated read-only copy, for another layer of hacker cracker safety).

In my mind, I envision the same sort of powerful reporting interface as those used by applications like Google Analytics. Date ranges, comparisons, graphs, charts... All would be quite welcome in a world of financial data. And the best part of such a system would be the near-obsolescence of paper reports. The dynamic charting and date-range selection capabilities of the Web application would surely prove more useful than fixed paper reports.

Of course, there would be Luddites intent upon keeping their paper-based reports. Hence "near-obsolescence" rather than just "obsolescence." There would have to be an option to receive paper reports, in recognition of the views held by the few people who would want them. But that option should be turned off by default. Anyone who wanted paper mailed to them would simply call the agency and say so. It could even be a question when new customers sign up. But again, the default would be "no." As far as I can tell, most people don't use the paper, so why waste it?

It's probably obvious that this is a half-baked idea. I haven't got any interface mock-ups, server-side code (say, PHP or Python), SQL, or anything else. All I have is an ideal, that paper-based reports be sent only -- and only -- to the people who want and use them.

Monday, March 24, 2008

StartClock: Save Taskbar Space

You know how Windows has that clock in the taskbar? It takes up a good 50 pixels on my screen, and that's valuable space where taskbar buttons or tray icons could be. And I know exactly what that big (green) button in the lower left does, so why do I need it to say "start" all the time?

With the freeware StartClock, you can change this:



to this:



What it does, through some AutoHotKey magic, is put a clock where your computer's taskbar usually says "start". The Start button functions normally; it just has numbers in the place where letters usually call home. That means you can turn off the clock in the taskbar (right-click the taskbar, click Properties, and uncheck the "Show the clock" box) and enjoy a few more pixels for your buttons and tray icons.

While the program does exactly what it is supposed to do, and is highly customizable (to the point of displaying a fake time, hard-coded, or even alternate text -- prank time!), it takes up about 6 MB of RAM, which might be a bit much if your system is tight on resources. My mom's computer has only 384 MB of RAM, so I would be hesitant to put too many tweaks on that PC, but my GB has enough room for them. Unless I try to play games, which I don't do much.

Besides the seemingly large amount of memory the program takes up, it also makes the Start button bigger, putting back some of the pixels you gain by hiding the default clock. The savings are definitely less than 50 pixels. I may or may not keep it around for very long, but it's still a neat idea. I quit using the program right after I wrote this post, actually. You might find it useful, but to me it's kind of clunky. It made for good GIMP experience, though.

If you're interested in downloading StartClock and taking it for a spin, head over to DonationCoder. Check out some of the other programs while you're there; that one page has a bunch of other maybe-useful apps to pick from. Though they might all end up being as iffy as StartClock; I haven't tried them all, by any measure.

Sunday, March 23, 2008

On Anecdotes and Other Stories

In my family (yes, I'm digressing from tech-talk again), I notice a very disturbing trend. The elder members, meaning those over the age of 55 or 60, tend to have incredible stories to tell that only come out when there is company around. When it's just us, the spontaneity is just not there to generate the proper atmosphere for storytelling.

That isn't inherently a bad thing. If we were to tell stories to each other left and right, our supply of anecdotes would begin to run low pretty quickly. But think about the implications of having company around while you're delivering a narrative version of some incredible or unlikely part of your life.

First off, you probably won't be in any position to record the story. At our house, at least, company usually comes as a package with dinner (when we have company over), which means everyone is busy beforehand preparing food, setting the table, and generally primping the house to look good. (One recent day of company, March 21, had me snowblowing the driveway while it was still snowing. Just an example of what we do for our guests, some of which I find equally illogical. But I'll save that for another post.)

In the midst of all this preparation, how likely is it that someone will have an audio recorder handy, to capture the stories that may, after all, be told only this once? In my experience, the probability has been zero. It's the same as having a camera available at all times, really. In a family of people whose experiences cover the world, recording the stories, somehow, some way, would seem to be a necessity.

Actually, audio recordings aren't necessarily the best way to keep stories. I mean, think about it. How searchable are audio files? Sure, you can name them and give them title attributes; but the full text is unavailable. And what of deaf people?

The best way to record these incredible anecdotes would be to have their tellers type them out, or speak them into a microphone. A problem arises from this approach, as well: Those whose stories we are interested in -- whose lives contain innumerable instances of incredible happenings -- find any attempts to record their history a waste of time. They simply have too many other things clamoring for attention. Why do taxes when you could be recording your life's history? Taxes only matter while the government still knows you're alive; your stories could inspire countless individuals decades into the future!

My point is not that taxes aren't important -- you can go to jail for not paying them, after all -- but that people whose life experiences deviate from the norm in even the slightest way have important stories to tell and preserve.

One way I could do something about this trend toward narrative oblivion would be to acquire a digital voice recorder and carry it with me everywhere, much as I do my camera. Should a story start up, I'd simply whip out the little device and begin recording. I would simply spend part of my free time each week transcribing stories from the recordings (editing out repeated words, pauses, etc.).

While the practice would soon require the expansion of my data storage capacity, I am already in need of such action due to the large number of photographs I take. It's no skin off my nose to use a few more megabytes a week from the 500 GB drive I am planning to buy. (It might even be 1 TB, actually -- I haven't fully decided; prices are hard to find for such large-capacity storage devices. Whatever I buy will definitely be RAID 1, though, for hardware redundancy.)

As I said, the ideal solution would be for the stories' protagonists to simply write them down, in one way or another. Failing that, I must take the audio-based route and transcribe the stories myself, because every time I suggest writing down their stories, the family members to whom I speak about it say it's not worth the time.

Like hell it's not.

People (often these same people) say blogging is a waste of time, too. I know otherwise. Some of the posts I have made to this site see dozens of visits a week, accounting for more traffic than my homepage. Posts like the one I wrote about Google Browser Sync's incredibly bad performance, my posted instructions to fix the problem with Tab Mix Plus and the new Gmail, and (though I am not a Mac guy) my news summary of the Wi-Fi problems encountered in Mac OS X 10.5 (Leopard). These posts have obviously made a difference in the world of computer users. However small, it is still significant. That is why I insist on trying to preserve the incredible stories of my elder family members.

I don't expect instant turnaround, but hopefully some of the people I am referring to, indirectly, in this post will see why I want to record their experiences. Perhaps I am a digital pack rat, but even pack rats have good reasons for keeping all their odds and ends. Just as I have good reasons for wanting to keep the stories alive.

Saturday, March 22, 2008

WOT Going 3.0

WOT, a website-rating (think McAfee SiteAdvisor, not StumbleUpon) browser extension available for Firefox and Internet Explorer, will soon be testing another new version. I was sent an invitation to test their last release before it was launched, and the upgrades there were pretty substantial.

WOT 3.0 will include comments and a plethora of details on sites rated using the add-on, including categorization, popularity, server location (country), and a direct WHOIS lookup link. It sounds even more impressive than the last upgrade. I'm excited!

I'm not sure how I found the WOT site, but when I first started using the extension, all it did was rate sites on four different categories (trustworthiness, privacy, child safety, and vendor reliability). Since then, it has expanded exponentially. The features introduced in the upcoming upgrade appear to rival, and may even surpass, McAfee SiteAdvisor, which is the extension I used before WOT (and dropped in large part because of the increased marketing after the McAfee acquisition).

I'm going to have to start rating sites some more; they say the top 500 most active users will be given the opportunity to test a pre-release version of the new WOT in the next few weeks. My site-rating activity has dropped recently, partly because I'm not visiting as many new sites and partly because I'm usually focused on something besides website ratings. I need to change that mindset.

Update (04/20): I was invited to beta-test on April 15, and I've just published my review of the new WOT 3.0 today.

Friday, March 21, 2008

Google Analytics Benchmarking Active

If you follow the other Google blogs such as Google Operating System or Google Blogoscoped, you'll probably already know that Google Analytics' Benchmarking feature is finally showing data. Heck, the Google Analytics Blog even announced this one. I thought I'd blog my take on it anyway, just to see how my opinion differs from those other people.

At first glance, the benchmarking data is cool. I haven't had a chance to probe every nook and cranny yet, but it looks like it is can be quite useful. That is, if your site fits into one of the categories provided.

I looked in the category list, and it didn't appear to have a "blog" category for me to select. Analytics just compared me with all other sites of similar size, however that's measured. All I know is, I have red marks on all metrics except New Visitors. I don't get as much traffic here as I'd like; hopefully that will change eventually and I'll have more green.

For those who are curious about what the benchmarking looks like, here's a screenshot I made from the data available for this site:



Like I said, red marks everywhere. If you didn't figure it out, the blue line is your site and the gray line is all the averaged data from other, similar sites. How about getting more of those blue lines above the gray ones? I'd certainly like that. I have a high bounce rate, low visits/pageviews, and a low pages/visit average. Must figure out how to grow this site... But I knew that without the benchmarks, didn't I? (Yes, I did.) Whatever, they still look cool.

Thursday, March 20, 2008

Remember The Milk's Flawed Sharing Model

Two days ago, I discovered my first glaring error in Remember The Milk. Two days ago, my mom signed up for the service. Two days ago, I shared a few lists with her as an experiment. Two days ago, I had to come up with a good hack, and fast.

That hack had to do one thing, and do it well: It had to get the shared lists out of her account. In other words, it had to unshare those lists.

The Dissertation

One would think that the ability to stop sharing something is an integral part of the sharing model. If you share something with a person, and then decide you no longer want them to have access to that data, you should be able to lock them out. True? Apparently not, even though Google Calendar and Google Docs (two examples that I use) both provide sharing and both let you remove other users from the sharing list.

Remember The Milk thinks that once you share something, the other person should have access to it for the rest of time, regardless of what in your relationship has changed:

Once you have shared a list with a contact, you cannot 'unshare' it.

The reason for this is that your contact may have added or changed some of the tasks, and 'unsharing' the list would mean that they would lose the changes they made.


Obviously my relationship with my mother won't change much, but their decision has serious implications for the business world. And that "reason" for not allowing unsharing is complete crap.

Granted, Remember The Milk is not to be used commercially. For people informally collaborating on a project, though, it's still an issue. One member may decide to leave, and the group task list would be stuck in their account. What's more, they retain the power to disrupt the work of the rest of the group, as all people with whom a list is shared are granted the ability to do everything the owner can.

That brings me to another point: Only a list's owner should be able to delete that list. The problems that arise when a user with whom a list is shared deletes the list have wide implications. For instance, none of the other members of that list can view the tasks any more. What's worse, though, is that the tasks are transferred to only one Inbox: that of the user who performs the deletion. Everyone else loses access to the tasks unless they are put in another list and reshared (or shared individually).

Granted, these flaws can be used for good, to transfer a list from one account to another (by modifying the unshare hack I'm about to share [and then unshare -- kidding!]). However, if list transfer is to be supported, it should be a separate feature, not based on hacks using existing flaws.

The Email

I wrote an email to Remember The Milk as well, explaining the problems I see in their sharing model. Here is what I sent to them:

Let me just state that as far as I'm concerned, the sharing feature is broken. Specifically, the fact that tasks and lists cannot be un-shared once contacts have been granted access is absurd. There is no way that everybody will always share something with the right person, or that the other people will be part of the project associate with that list forever.

Say a pair of people who have started a small company together decide to bring in a third person temporarily. They share with that person a list of tasks related to the business. Then the third person quits their job and moves on. That third person should no longer have access to the business's task list. While it could be said that it is up to the user to make sure they avoid situations like this, the real world is never perfect.

Where possible, software should smooth over the experience of using the service, and this includes un-sharing. For now, my mother will have three useless lists in her task screen that were meant to be published, not shared. I have tried deleting contacts on both ends, only succeeding in making the RTM interface say the lists are not shared even though they appear in her view. Deleting the list and then restoring it also failed.

For what it's worth, my RTM username is voyagerfan5761 and my mother's is <scrubbed>. If there's a way you guys can clear the sharing status between the two accounts, that would be wonderful! Implementing un-sharing would also be absolutely priceless (and might earn you guys another Pro subscriber, wink wink :D ).

Anyway, I'm done with my rant. Please keep me apprised of the status of this ticket.

Voyagerfan5761


Note that the letter was written before I figured out that the word 'unshare' was supposed to remain unhyphenated. Oh well. I'll remain true to the quote rather than edit myself.

The Hack

Anyway, I did manage to hack together a workaround to unshare a list in RTM. Here are the steps:

1. Clear your Inbox list by moving everything to a temporary new list to make room for Step 2

2. Take the shared list you want to unshare and delete it in the Settings







3. Create a new list with the same name



4. Go to the Inbox and move all incomplete and complete tasks to the new identically-named list



5. Delete the temporary list created to hold any Inbox tasks (which moves them back to the Inbox list)

6. With that, you should have your list intact, and with you as its only user again



The contacts with whom the list was shared will have to delete it from their accounts themselves, unless they have not yet accepted the list. (I consider it another flaw, BTW, that deleted lists are merely emptied for users with whom they are shared, rather than being removed completely.)

Do not -- repeat: do not -- delete contacts in an effort to unshare something. That will only make matters worse, as RTM will continue sharing but not display the fact (shared items will display as unshared even though the other users(s) still ha(s|ve) access). However, even if you have, this hack will still work. I know because deleting contacts was the first thing I tried.

Conclusion

So anyway, this is a pretty complete write-up of all the things I found wrong with RTM's sharing model. I just hope someone (Emily and/or Omar, preferably) takes notice of this post...

Wednesday, March 19, 2008

Google Maps Becomes Wiki-Like

Readers of my blog should know by now that I stand by the wiki, a kind of freely-editable website. Google got into the wiki business, sort of, when they acquired Jot (not Jott; that's something else) and developed it -- after many months -- into Google Sites. Which is a wiki-that's-not-a-wiki.

Well, recently Google Maps got some updates that make it also a wiki-that's-not-a-wiki. Users from the United States, New Zealand, and Australia can now add, edit, and delete places in Google Maps. (That Google Operating System post has tons of screenshots, so just head over there to see them if you're interested. It saves me the time it would take to make my own. :-D )

The edit function has been around for a while, but was limited to placemark location. Now the details can be edited as well, including address, name, and all the works.

In addition, places can now be added and removed, though removals require Google's approval.

So what's with all this community-based stuff? I have a feeling Google is capitalizing on the same forces that helped build sites like Wikipedia. That is, the wisdom of the crowds. Most people jump at the chance to add a missing entry to a database they use; it makes them more likely to use that resource if others can add missing information.

I already use Google Maps over MapQuest or Windows Live Local (I think that's what Microsoft calls it now). All these new things just keep me coming back. One of these days I'll have to try adding a missing listing just to see how it works. That just means I have to find one.

Tuesday, March 18, 2008

Remember The Milk's New "Pro Tester Program"

I just found out not too long ago that Remember The Milk has launched a new sort of alpha-testing service where users can test out unreleased features before they are finished, polished, and released to everyone. Admittedly, it's only been available since the middle of my stay in Milwaukee, so I wouldn't really have found out about it, but now that I'm back I have some things to say about the program.

First, a little background. The Pro Tester Program (more on that first word in a moment) is intended to aid in the testing of new features on multiple and diverse browser/platform combinations. From the RTM blog entry:

PTP members will be given pre-release versions of new RTM features to try out, and will be invited to provide feedback and help us by reporting any problems encountered with said new features.


That sounds exciting. It's a lot like Google's limited roll-outs that happen from time to time as they implement new features in services like Gmail or Google Docs. A few users will see the new features, and nobody else will. The difference? First, it's not random. Users sign up for this testing. Second, the participation requirements are actually published (my emphasis):

  1. An RTM Pro account (the "Pro" bit of "Pro Tester Program").
  2. The ability to keep a secret (you'll be testing features that aren't available to the public yet!).


Wait a minute, I don't pay to use the service so I can't test out new things for you? I'm not completely against paying for RTM -- it is, after all, a wonderful service -- but it seems like I should be getting paid to do your testing for you. After all, users who sign up for the testing service will be saving the RTM team work. Why is a Pro account required?

Granted, RTM needs to make money somehow, and adding Pro-only services as incentive to buy a subscription is a valid business model. If I ever have need of synchronization with my Windows Mobile device, I'll pay for it. If I ever want to get a special iPhone/iPod Touch interface, I'll pay for it. If I'm going to be alpha-testing new features for a beta service, I won't pay for it. I'm sorry RTM. I love testing, but paying for the privilege of saving you guys work just doesn't cut it for me. Make it accessible to all your users and we'll talk.

Oh yeah, my notes for this post (stored, ironically, in a Remember The Milk list) say that your new program is similar to Google's Trusted Tester program as well. Similar because people who apply get access to pre-release features. Dissimilar because you have to pay to get at it (and also because you don't have to be blood with someone at RTM to do so).

So, while this new service has increased the probability of my buying a Pro account, it has definitely diminished my respect for how RTM structures its business. And no, I won't buy a Pro account now, just on the principles I have outlined here. I will not pay for the privilege of saving someone else effort; they should be paying me, or at least not charging me money.

RTM: Try allowing everyone to participate in the Pro Tester Program. You'll probably get more feedback, which means less time overall pushing out updates. Please?

Monday, March 17, 2008

GIMP Script to Make SW Robotics Thumbs

After making a whole bunch of thumbnail images for the member and mentor profiles on the Southwest Robotics site this past season, I finally decided last night to automate the process with a GIMP script, in preparation for next year's new people.

So, what does it do? It takes the image it is run on, scales it (gimp-image-scale) to 120px x 90px, and utilizes script-fu-drop-shadow to add a correctly-positioned drop-shadow. (Correctly is, in this case, the same as all the other images I did manually.)

It's a rather simple script; here's the code:


; Script-Fu script to automatically
; resize and drop-shadow files for
; SWR profile shots
; IMPORTANT!
; Image must be EXACTLY 4:3 aspect
; ratio or it will be distorted.
; This script does not measure the
; image or do any cropping; it is
; up to you to crop the image properly.

(define (swr-make-profile-thumb img drawable)
(gimp-image-scale img 120 90)
(script-fu-drop-shadow img drawable 4 4 8 '(0 0 0) 80 TRUE)
)

(script-fu-register "swr-make-profile-thumb"
"Make SWR Profile Thumb"
"Scales the image to 120x90 and adds an appropriate drop-shadow"
"Voyagerfan5761"
"Voyagerfan5761"
"March 17, 2008"
"RGB*, GRAY*"
SF-IMAGE "Image" 0
SF-DRAWABLE "Layer" 0)
(script-fu-menu-register "swr-make-profile-thumb"
"<Image>/Script-Fu/SWR")


Note: The original has my real name and an email address, but I have stripped them out for purposes of posting this in view of Web crawlers.

Input images must be in 4:3 aspect ratio to avoid distortion. I currently have no validation for aspect ratio, nor am I certain that it's even possible to check such things, but if it is I will add it at some point. As can be seen from two of my updates last night on Twitter, I only just learned Script-Fu Scheme less than 10 hours ago.

The only real problem I had was developing the script-fu-drop-shadow line, which kept choking when I tried to set an interactivity mode. I eventually figured out that that parameter wasn't required in 2.4 (or some other weird problem; I'm still not sure).

It's obviously not much, just a few basic lines of the stripped-down Scheme that GIMP uses. Nevertheless, if you use my script as a base for a released script of your own, I would appreciate an acknowledgment. A link back here along the lines of, "Based on a short script by Voyagerfan5761" would be sufficient (note the link on "Voyagerfan5761").

And of course, I'd appreciate any and all comments on this script, be they pointing out a better way to do things, something I missed, a suggestion for validating the aspect ratio (hey, it never hurts to ask). Or even a note that you based a script of your own off of mine. That's highly unlikely, considering I didn't even copy any code examples for this one. (OK, I lifted and modified the registration function from a GIMP tutorial page at GIMP's website, but that's all.)

Sunday, March 16, 2008

FIRST Wisconsin Regional Results for Team 2129

Note to visitors on this post: If you got here while searching for the full competition results, I apologize (they can be found at the FIRST website). My original title was rather general, and I have changed it to be more specific. Search engines might not pick up the title change for a while, though, so again I am sorry if you got here in error.

Well, I got back from Milwaukee around 00:20 last night (this morning?) and went to bed within an hour. Among other things, the bus taught me that manufacturers don't pay enough attention to uniform seat spacing. There was one row with an extra several inches of legroom. Sounds like airliners to me. (Oh wait, airlines are today's bus companies. Duh. No wonder the seating arrangements are similar.)

Anyway, yesterday's competition was, for us, only a half day. Ultraviolet, FRC Team 2129, finished at #48, out of the 60 teams in attendance at the Wisconsin Regional Competition, with three wins and five losses. We were done before lunch.

We sat through the top eight teams picking alliance partners for the final rounds, watching our team number stay in the Available column through choice after choice. Eventually we ended up as an alternate for one of the teams (I've forgotten the number), meaning we would have competed in their place if their robot broke beyond reasonable repair, but of course that didn't happen.

The second half of the day, from about 13:00 to 17:00, was spent just watching the increasingly competitive finals between the last alliances. It wasn't a total waste of time because we got to watch a couple other Minnesota teams compete. Robohuskie (2574, St. Anthony Village High School), The Green Machine (1816, Edina High School), and Blue Twilight (2220, Eagan High School) all competed, as well as The Millerbots (2549) of Washburn High School.

Among other things, Team 2549 got into the upcoming nationals. In a way, it's a bit annoying, since we mentored them and this is their rookie year (2129 started in 2007). What's also interesting is the fact that, while our robot could do everything (make laps, push balls, carry balls, put them over the overpass, and place them up there at the end of the match), Washburn's robot had very few capabilities. It was much like the robot from Mukwonago, WI (930), which did only one thing: race around the track and make a lot of laps. Even with the annoyance of being beat out by a rookie team that got their start from us, the stigma is just our dented pride; all of us hope that Washburn does well in the nationals.

Moving on, I took a lot of photos at the events, going through, on average, a battery and a half per day for three days. I also have several video clips. I'm thinking about uploading a couple, perhaps one of the robot in qualifications and one of our team goofing off in the stands, and embedding them here in a later post so as to provide some insight into what happens at those competitions. All the FIRST video sites I've seen have yet to get footage of this week's regional. Heck, even FIRST's own site is lacking alliance details for the latter half of the final rounds. There's a huge amount of data to be uploaded, for everyone involved, so some delays are expected.

On another note, I'll now be back on my normal current-stuff posting schedule. Even though I brought my computer, I didn't use it very much. Time actually spent at the hotel was very limited, and most of it was spent sleeping. We were up at 06:00 every morning and didn't get back until between 20:00 and 22:00 every night. Sleep usually came around 23:30 or 00:00. Email and a bit of Twitter (through Twhirl) was about all I used the computer for.

Over the next couple days I'll get caught up on blogging about stuff that's happened since this past Monday. There are some good Google things in there, as soon as I can get more info on them and write up some semblance of a post on each.

Hope you've enjoyed following my (sparse) updates on the FIRST Wisconsin Regional, and hopefully we'll do better next year. (For reference, we made it to the finals last year. As I said yesterday, we had some bad luck this year.) Meanwhile, there should be some professionally produced photos and videos at the Wisconsin Regional website at some point.

Update (20:17): Changed the title to make the specificity of this coverage more obvious and added a note to that effect. Since I only really talk about Team 2129, the one I'm on, I figure I should say that in the title. That and I see visits from Milwaukee who probably found this in a search for the general results, which can be had at the FIRST website.

Saturday, March 15, 2008

Milwaukee Status

Yesterday we had the qualifying matches for the FIRST Milwaukee Regional Competition, which I went to with the Southwest Robotics Team. The day before yesterday was practice matches, as an opportunity to work out bugs on the official arena before the actual competition.

The good news is, our robot was moving up in the rankings. The bad news is, it completely underperformed what I, and many of my teammates, expected.

We battled mechanical problems (our lift prongs got bent on Thursday) and yesterday we couldn't seem to do anything with the lift arm until our final match of the day. The robot didn't tip over (which it did do twice in practice competitions a couple weeks ago), but it just wasn't performing at the level it had been in practice.

Today we'll see how it does, and then go back to Minnesota. We get home around midnight, so I probably won't be blogging about the results at all until Sunday morning. Sleep this week has been a precious thing; each day is a lot of walking around the large arena. Your feet and eyes start to ache and burn (respectively) starting around 16:00, and when you get back to the hotel you just want to go to sleep.

Actually, you want to go to sleep right when the FIRST officials say that the day is done, but there are other (non-robotics-related) activities to do and no time to nap.

I truly have no idea what we'll be doing this morning, but we're about to load the bus with our luggage and walk to the arena, so it shouldn't be long before I find out. Obviously we'll be running the robot some more. Er, I mean "hopefully," not "obviously." :D

Yes, I know I said I would be posting canned material through Sunday, but I ran out of stuff to post, and I ended up bringing my computer. Plus there's been time to write. (Precious little, but time enough.) So sue me for lying; at the time I wrote my last "current" post, I thought I would be offline.

In light of that, consider this a sort of "bonus" post. Now I have to think of something to publish Sunday morning...

Friday, March 14, 2008

Internet Explorer Upgrade Nags

Fortunately, Microsoft hasn't pushed out a required upgrade to Internet Explorer 7 for Windows XP yet. They're definitely trying to get people to upgrade, though. A few times now, I've had the Internet Explorer download page from Microsoft.com open on me, instead of my homepage (iGoogle over HTTPS).

At the moment, it says, "Upgrade with confidence," right under the first heading (normal text, below "Internet Explorer 7 downloads"). Sure I have confidence in how much better IE7's rendering engine is, but I don't like the new interface or their implementation of tabs. That's why I still use Internet Explorer 6. I know that the sites I design look better in IE7 than in IE6 (though I've fixed a lot of bugs; and in all honesty they were designed with better CSS support in mind, i.e. Firefox 2).

So it bugs me when IE6 tries to get me to dump it for its younger sibling. All the sites I use in IE work fine in version 6, and in fact I was told that version 7 is unsupported by the most critical sites of all: my school's. Being nagged to upgrade when I have been told not to is annoying.

Microsoft: If you're going to offer an optional software upgrade, offer it in Windows Update, and/or a download. Don't bug users who can't (or shouldn't) do anything to change what version they use.

Update (03/16): Just got another one when I opened IE6 today. Grr...

Thursday, March 13, 2008

Windows Update Reboot Prompts

No, I wasn't recently annoyed by one of these, but I thought I'd mention them while I'm out of town and unable to keep up on the latest news.

So every PC user has probably seen a restart prompt by now. Windows pops them up every ten minutes after installing updates, trying to get you to reboot the computer and apply the new code. What Windows doesn't take into account is the fact that you might be in the middle of something that takes more than ten minutes.

Granted, there are workarounds (like using the Group Policy Editor), but it's an unnecessary annoyance for power users. While the system policies are an OK way to turn them off, a simple checkbox in the Windows Update control panel wouldn't hurt for those of us using, say, Windows XP Home Edition (which lacks the policy editor) or who just don't want to bother mucking with convoluted settings.

So yeah, I don't have an image because I wasn't thinking about a post like this the last time I got updates, and I couldn't find a way to force the dialog (there must be a way to fool Windows into thinking it just installed updates; if you know one, a comment would be useful :-). I do have a screenshot for my next annoyance, though.

There, that was a nice, not-so-graceful segue into the subject of the dialog where a choice isn't really a choice.

Windows has a very annoying way of severely limiting your control over the system's update process if you're a limited user. If you use a computer managed by someone else -- say school, company, or whatever -- you might have seen this dialog:



It gives you an amount of time (in this case it was five minutes) after which the system will reboot whether you're ready or not. What really burns me is how it has both "Restart Now" and "Restart Later" buttons, but the latter button is disabled. It's probably there because the same function is used for creating all the restart prompts with a conditional to enable/disable the timer. But that's getting pretty technical.

Suffice it to say that not having the option to stop the timer and finish what I'm doing if I need an hour is a large annoyance. (That screenshot was taken several months ago for submission to This Is Broken, now shut down, but was never acknowledged and has been sitting in my email account ever since.)

Wednesday, March 12, 2008

Going Out of Town (FIRST Milwaukee Regional)

As you all probably know by now, I am a member of the wonderful (am I biased?) Southwest High Robotics Team, whose website I designed. Well, we're going to Milwaukee this afternoon, and we'll be back late Saturday night (CDT, of course). It is quite interesting that the team captain hasn't posted about the trip yet on the team blog... But I digress.

What that means for this site is that I'll be publishing stuff I wrote yesterday (and this morning) and scheduled using that wonderful feature that Blogger's still testing (and recently fixed [wait, not fixed-- yes, it's fixed] a bug in). The posts came from ideas that have been sitting in my idea bin (actually a list in Remember The Milk) for a while but have been overshadowed by current events and updates from Google and others.

So there should be no time-sensitive posts made between today and Sunday, March 16. (If anything more than one post a day is published, someone's hacked my account, which would really be a bummer.) I've done my best to make decent rants about things that Microsoft (for example) did in its software that I find annoying.

The only problem is that I've had less time to write because I've been trying to get ahead caught up in school before the trip. My next several posts might seem a bit thrown-together because of that, and I'm sorry for it. I should be back to my normal quality once things settle down. (I wrote two of the posts yesterday morning in about 45 minutes, which is usually about how long it takes to do one entry.)

If you're also a member of a FIRST robotics team and will be going to the Milwaukee Regional this week, look for me, the guy in a purple shirt walking around with a small camera (I also do photography for the team). I don't expect to find any readers there, but hey, who knows?

So enjoy the canned posts, and I'll catch up on any news that happens next week. If I get the chance, I will try to post Twitter updates. Follow me on FriendFeed to get all my activity (including here, Twitter, YouTube, StumbleUpon, Digg, etc.) or on Twitter for just my tweets.

Tuesday, March 11, 2008

Debug Constants vs. Secret GET Parameters

A while back I commented on one of possible248's posts on CodingExperiments.com that secret GET parameters aren't the best way to do debugging. The problem is that if someone discovers your "secret" parameter (which most people would probably set to just debug=1), they can get all sorts of information about your site's underlying code structure.

What I like to do is have debug mode switched in the source code itself, say in a global include file that defines constants and variables for the entire site. (MediaWiki's LocalSettings.php is a good example.) Since it's usually wise to do your development on a different set of files (at least, if not on your own machine), you can make changes and set debug mode on while coding, and then turn it off before uploading the code to the live site. There's no risk of someone discovering a hidden parameter, and you use the same basic if(debug){print debug stuff} code that you would otherwise.

It's not necessarily something you would call a "best practice"; it's just the sort of thing that you want to think about before making the decision to use GET params or constants. If security is important, you should stick with things that can only be switched by modifying the source code. Barring anyone hacking your server, everything would be safe from GET snoopers.

This is just some food for thought.

Monday, March 10, 2008

Musical in the Food Court!

This post digresses from my usual technobabble, but it's just too funny and creative (not to mention awesome!) to pass up.

Improv Everywhere is a New York City-based group of people started in 2001. What they do is usually totally off-the-wall and almost always funny. Their blog's tagline: "We Cause Scenes". The mission I'm featuring today, though, was done in Los Angeles, CA. (There is now a global Improv Everywhere site on Ning as well, but this mission was performed before that was launched.)

I like the idea of Improv Everywhere because I am also into improvisational performance. I don't usually get into the complexity they do, nor do I pull off stunts in public places -- in fact my improv is usually part of some exercise in theater classes. I do like to do things like randomly start singing in the middle of a mall, and that's what their latest mission was.

Posted to ImprovEverywhere.com late last night, "Food Court Musical" has already drawn a ton of comments. The story is, well... Watch the video first.



I don't know who came up with the idea to sing a song about needing some napkins, but this is just too funny. The way it starts with one person and escalates into an entire group is really great.

One of the best parts, I think, is the security guard. He comes in and everyone thinks he's about to break up the performance, when suddenly he starts to sing as well.

The title sounds to me like a cross between the two Disney films High School Musical and Full Court Miracle. It really works, I think. The choreography even reminds me of some stuff from HSM.

As described further on the mission's page, Improv Everywhere actually got permission from the mall to pull off this show. That's something they don't usually do, but in this case it made for a better performance. Having people dressed in the same uniforms as mall personnel and putting one behind the counter of Hot Dog on a Stick definitely made it seem more real. (The other food court businesses had no idea what was going to happen.)

Anyway, I won't lift all the notes from Improv Everywhere's mission page, even if I am rewording it. They deserve traffic for this one, so head on over there for more commentary and the photos.

Before you go, however, I would appreciate a comment here if you like this sort of post, or if you don't. I need feedback so I know what to write more about. ;-)

Update (13:28): Reduced embed size; it was too big for the content column. It should no longer overflow into the sidebar.

Sunday, March 09, 2008

Funny Randomization: Google: Web 2.0?

Here's a randomization from google.isyournewbicycle.com that is obviously untrue, this time about Google (rather than Microsoft):



That's really funny, considering that Gmail was one of the first major Web applications to use AJAX extensively. Google's Calendar, Reader, and Docs services (among many others) also use AJAX as a fundamental part of their inner workings, so it's even funnier in that case.

Saturday, March 08, 2008

GOOG-411 Expands to Canada, Possibly Elsewhere

I've heard too much about this subject to dismiss it as a rumor. I've read posts from multiple sources that GOOG-411, Google's free 411 service for businesses, is now available in Canada. Garett Rogers, a blogger at ZDNet who lives in Canada, confirms that the service does indeed work in his city.

Haochi of Googlified says that Google has not yet updated the page telling where GOOG-411 works, and my testing confirms that it still says the service is only available in the United States for U.S. business listings.

So, this doesn't really affect me because I live in the United States, but it's definitely a sign that the service has reached a certain level of maturity and reliability. Google doesn't usually release services in other countries unless the English (US) version has performed sufficiently well in its initial release in the United States.

I have a feeling we can look forward to GOOG-411 eventually expanding to other areas where Google has business data available. I think it shares the same database as Google Maps, but I don't have a source for that.

Friday, March 07, 2008

Blogger Team Says Permalink Issues With Scheduled Posts Fixed

User Belcarnen noticed late March 4th that the permalink issues seemed to have been fixed. Yesterday morning, Pete was back with another note saying that the problems should be fixed this time. In his words:

As Belcarnen noticed, the permalink problem should be sorted out now.

Jason assures me that the correct code has been pushed to all servers and all is well.

We do have reports of scheduled FTP publishing now not working, which we're looking into. We wanted to get the permalink thing sorted out first, though.


So there are issues with scheduled posts not working on FTP-based blogs now. I think that was actually a known issue in the original release... Oh, yep, it was the last bullet point of the notes at the end:

  • In some testing we’ve found that schedule posting is not working for some FTP blogs. If you run across this, please let us know in the comments so we can try to track it down.


Going back to Pete's comment, I'm going to assume that this Jason he mentions is one and the same with the Jason Wilson who originally posted the announcement. If the correct code has been pushed out to all the Blogger servers, there shouldn't be any more issues with permalinks.

I'll be running my own test using this post, scheduling it to publish an hour or two after this writing, to see if my blog's server is indeed fixed. All I need is a way to easily predict the permalink I'll get, so I can update past posts before the scheduled post publishes.

If you know of a reliable way to predict the permalink, let me know; I could really use it, as I run into this problem a lot! I know my domain name, so I won't worry about that, but the part after the slash is in the format /yyyy/mm/word-word-word-word-word.html. What I need is a way to predict the word-word-word-word-word part, without briefly publishing the post first. That way I can change the title before publishing and the link will change to accommodate the new title.

Thursday, March 06, 2008

Google Reader Adds "Most Obscure" Tab to Trends, More Reading Space

Continuing from the data available on the Official Google Reader Blog, and so as not to make my last post about the "show details" links too long, here are the other changes to Google Reader.

Aside from adding more languages (which is ongoing), the Reader team managed to find an extra 17 pixels, vertically, for content. They also added a small feature that shows which of your feeds have the fewest (Google) subscribers:



It's actually kind of neat to see how many sites I've subscribed to that have fewer than 10 readers on Google (right now I have five). Notice how my friend i80and's blog is in there. Will somebody please subscribe to him? (If he asks how you found his site, tell him it was through Google. ;-) While you're at it, if this site is interesting to you, why not subscribe here, too?

OK, enough feed-pushing, I'll stop.

So that's the latest batch of stuff from the Reader team. I wonder what they're working on next... Authenticated feeds, I hope. Maybe better Gears support so images will also be cached...

Google Reader Adds "Show Details" for Quick Stats

A while back, the Reader team added pop-up, tooltip-like windows to the timestamps of posts in the reading pane to show the times that each post was published and added to Reader's cache. Now there are easily-accessible statistics for each feed, visible after clicking a new "show details" link:



*Click*




The bar that pops up contains info on the average number of posts per week for that feed and the number of Google subscribers. In these screenshots, I've included a sample of what the stats look like for the feed from xkcd.com (a webcomic you may remember, as I wrote a list of my personal favorites a while back).

xkcd's main feed has about 27,500 subscribers as I write this, and posts approximately 3.3 items per week, which is on par with the stated update days (Monday, Wednesday, Friday).

The write-up on this and a couple other new features actually showed up on the Official Google Reader Blog as I was composing this post. I swear, when I started writing, this was an unannounced new feature. But I will give credit where credit is due, as I wouldn't have found out about the increased reading area and the new "Most obscure" tab in Trends.

Wednesday, March 05, 2008

Google Analytics Rolling Out Benchmarking and Sharing Settings

Google Analytics makes a great way to track how your site is doing, visitor-wise, and see what content attracts the most eyeballs. If you don't have a lot of experience in the Web analytics world, like myself (I won't lie, I'm a n00b), you'll probably find these upcoming features useful.

What they'll do is allow you to see how other sites in your vertical market perform. (Your "vertical market", or "vertical", is your field, e.g. e-commerce, blog, etc. I think. Like I said, I'm a n00b.) To get the option, those of us with higher account numbers will have to wait until Google rolls out the options to our accounts.

Once you do have the option, you'll have to enable it by allowing Google access to your Analytics data for purposes of analysis. You can't get benchmarking data without sharing your own statistics, and it may be up to two weeks before you'll be able to see any benchmarking data at all. Nevertheless, it sounds like a useful feature that I will be glad to have once it's enabled on my account.

I won't go into heavy detail because the Google Analytics Blog already has it well-covered. I will say that you have to share all data in your account; you can't enable selected website profiles. That seems like an important point.

Also, there are granular data sharing options, so you can share your data with everyone (anonymously) or just with Google, allowing upcoming enhanced services to work with Google Analytics and become integrated in one interface.

For more specific details and a screenshot, hop over to the Analytics blog; I'm just writing a summary and brief opinion here. No sense writing again what has already been said by other people who understand the new stuff a whole lot better than I. ;-)

Tuesday, March 04, 2008

Blogger Scheduled Post Permalink Issues Actually Not Resolved

The Blogger in Draft team's announcement that the permalink issues with scheduled posts were fixed is incorrect. One of the developers, Pete, left a comment today acknowledging that the problem still exists, and advising users that there will be an update (whether comment, new post, or edit) when it is really fixed. His note came in response to comments from another user and myself regarding the bug's continued existence.

That permalink problem is really a bummer, since it messes up the site feed and the homepage. To fix it, all one has to do is go into the admin section and republish the post, but such a workaround defeats the convenience of scheduled posts.

Here's hoping that the problem will be fixed, for real, in a timely manner; good luck and best wishes to the Blogger development team!

Update (03/07): Reported as fixed.

Monday, March 03, 2008

WordPress' Impressive Import Functions

This post has to do with an experiment I conducted last night with the WordPress installation at CodingExperiments.com, one of the other sites I write for. In the true spirit of that site, which is experimenting with code and all things computer-related, one or two visitors (who viewed the site or requested the feed in a window of about two minutes) may have noticed six extra posts briefly appear in the archives. That was my doing. (I did not, however, have anything to do with today's outage; that was possible248. :-D )

See, possible248 (that site's owner) granted myself and i80and (another friend and contributor to that site) admin status within WordPress (I'll have to talk to possible248 about his capitalization, as he writes it "Wordpress") yesterday, and I was exploring the options offered in the control panel. One of the tabs that intrigued me greatly was the import tab, which offers many options for importing content from other platforms:



That... is... so many platforms. Blogger being there is of particular interest, because I currently write for several blogs hosted on Blogger, including this one.

So, to explain why six posts appeared on this site briefly: I decided to give the import a try on the smallest blog.

The import was completed very quickly using Google Data APIs. After the import was complete, I was given the opportunity to change the author of posts. I must say I am very impressed with WordPress' ability to grab content from so many (13 plus itself) different sources.

It's also reassuring to know that, if I ever choose I no longer like Blogger and want to host my own site, it will be a piece of cake to move to WordPress (which is what I'd probably choose). The only question there is moving back to Blogger if I ever decide to do that. But both situations are far out in the future, reserved for when I have the time and money to maintain my own hosting.

Meanwhile, I consider the experiment a success. Go WordPress!

Sunday, March 02, 2008

Google Talk Actually Not Dead

Ionut says he's heard from a Google engineer that the downloadable Google Talk client will be updated sometime in the near future. We can expect a completely new version of the program, one that may (from screenshots posted on Google Operating System last year) include the ability to call phones. Supposedly that functionality has been available "since long" inside Google.

There hasn't been a Google Talk client update since it was upgraded to work with Windows Vista 14 months ago, and I've previously written about the new features introduced in Gmail Chat and the Talk Gadget not being added to the client. With this news from Ionut, though, I have renewed hope that I will soon see a reason to actually switch back to Google's client from Pidgin.

Nobody seems to have a list of new features, or even an estimated release date, but that's not unusual for Google; they seem to like keeping their release schedules private. But we will have a new version of the Google Talk client. It's just a matter of time.