Vacuum

The import-export problem and fragile Internet services

Here's the scenario. You and your team work hard to use a new piece of collaboration software that offers some unique features that are suited either to the work at hand or that appeal to your sense of novelty. The tool works great and you get into it, using some of the more complex parts of the system to great advantage. Then the system's owners send out the dreaded self-congratulatory message saying that they have been acquired by a big company, or some beleagured product manager announces that their system is being sunsetted. You need to move.

Moving turns out to be hard. Once upon a time when the only thing you needed to move was electronic mail, there were only a handful of email file formats, and though they were incompletely specified you could readily move among them. Moving between news readers was just a matter of reusing your .newsrc file. Life was simpler then (cue sad violins).

The situation now is much more complicated. If you are extremely lucky, the system you have has an export file that is the same as its import file, and someone has built a faithful clone of your original service for you to migrate to. That was the case in my move from Delicious to Pinboard. If you're keeping a weblog, though, things can get really unhappy really fast, and a friend to recently migrated from Posterous to WordPress went through gyrations that I'm not sure that I would have tolerated.

The fundamental problem, I suppose, is that the more innovating the service the less likely that its data is structured so simply that you can drop it in to someting else. Brand new user interfaces, the ones that attract traffic because of the ever present need for novelty, also have novel data structures. Watch out when getting into something new that you may find it easier to start than to stop.

Related articles

Twitter API changes cause obvious problems for developers

Posthumous Posterous: the importance of the exit strategy

Backup Your Online Life

Daily Tip: Export your feeds from Google Reader

Draft Is The Best Word Processor You Can Write With Today

How to Export Your RSS Feeds From Google Reader

The D2 Pad, a short review of an Android tablet

The boys got D2 pads (http://d2pad.info) for Hanukah. Here's a 250 word long review of this 7" Android tablet.

I've never owned an Android tablet of my own, so I don't have much to compare against. The experience so far recommends it as a handy adjunct to a full-size computer for some tasks, and it makes me more confident in my hunch that an Android phone is going to be my next phone.

We ran into problems charging one of the devices, but helpful tech support fixed it. The D2 pad would not charge with the provided wall adapter, but it did charge just fine using a USB cable. I can't explain why.

The devices don't support the Google Play store. I find myself mostly using the Amazon App Store, which appears to be a credible alternative. I'm sticking to free apps for the moment; the favorite so far is Swype, an alternative keyboard.

The camera on the device grainy even in good light, and it takes crappy photos in low light. It would be fine for a low resolution video chat.

The touch screen is surprisingly good for such an inexpensive device. On one of the two units, we found that there was a small defect where it appeared that one corner of the screen was being pressed down on continuously; that seems to have gone away.

I wish the device allowed true multiple user operations. #2 son is getting tired of having my Google calendar alerts show up on his screen.

UPDATE:

One of the pads now has a cracked screen, after someone got a bit frustrated with it. It still functions just fine. The funky defect on the corner of the screen never really did go away.

The Android operating system environment gets an enthusiastic thumbs up from the more vigorous user of the system, who is looking forward to the day when he gets better hardware to work with. He's showed me all sorts of fun things including an Android based remote control for Youtube.

 The boys are getting used to seeing Papa's Twitter show up from time to time on their screens, and #1 son is looking forward to the day when he's old enough to get his own Twitter account.

Related articles

multi user Android devices?

Which Android Tablet should you buy on a budget?-Tech Tuesday

Top 3 Tablets Of 2012

First look at the Chuwi V99 Android tablet with Retina Display

Android in 2013: Expectations and Predictions

The D2 Pad, a short review of an Android tablet

Prompting questions

For reference, here's some prompting questions that are showing up in my social media streams. These change from time to time as online sites try to cajole, coerce, trigger or prompt people to participate online.

Typepad wants me to share this post, and to "Enter a short message about your blog post to get people excited to come read it."

Facebook is asking me "What's the best thing that's happened to you today?" I'm sure that this is not the only question that Facebook asks, based on other people's reports, but I haven't yet seen a comprehensive list of the prompts.

Twitter is asking me to "Compose new Tweet..." 

And Google's Gmail is just directing me to "COMPOSE". Better to compose than to decompose, but looking at just the one word out of context makes it look very out of place. Email is of course much harder to write than a wall post or a tweet, since you have to come up with To, Cc, Bcc, Subject and email text and since you can go well beyond 140 characters.

"The business model is hamsters generating less money with each turn of the wheel."

The phrase is from Dave Winer, noting the path of Web 2.0 company growth and the difference between what the systems are worth to the owners vs. what they are worth to the early adopters:

The APIs are corporate APIs, the CMSes are silos, the business model is hamsters generating less money with each turn of the wheel.  

I can't say this is wrong, but it brings some regret for the time spent being the early hamster on the wheel and seeing whatever value I added to the likes of brand new systems being captured completely by those systems. It makes me want to start something brand new of my own - but then - you end up being captured by all of the tools you have to learn to bring it online.

Perhaps the best thing to think through this is that all new shiny online services have a lifespan. Some of them grow quickly and then die (remember Pownce? I thought not), some of them grow slowly and still live on somehow (remember Plurk?), others never get off the ground (a dozen tiny spam-overrun Twitter clones), and a tiny handful make it to Twitter size. You generally don't know when you're starting to talk to people through some new online system whether you are going to treat it as a long-term commitment to learning a complex system, or just some hands-on experimentation in something to be ignored and forgotten in a few weeks.

So you are a hamster, and you see a wheel. Do you get on to try it out, give yourself some exercise, or do you go to your workshop and build a better wheel?

Related articles

Twitter API changes cause obvious problems for developers

Hamsters Are Watching You

Google Analytics real time vs. mybloglog

We are writing to inform you that you have now been added to the beta for Google Analytics Real-Time! As we announced the Real-Time reports show you what is happening on your website as it happens.

Once upon a time, there was mybloglog. It was great, and then Yahoo bought it, gave it the big purple Yahoo squeeze of love, and then killed it.

Google now has a real time version of Google Analytics, which does some of the things that mybloglog once it. It tells you in real time what pages are being looked at, where people are coming from to look at them, and which keywords and referring sites are driving traffic. It scratches some of the same itch that mybloglog used to scratch, when there's some kind of burst of news and you want to know what's going on in real time and not on tomorrow's reporting.

mybloglog had something that GA doesn't have and really hasn't had: tracking of clicks not to the site you're looking at, but to something out there in the world that you linked to. That was great to use - and something that made it easy to see that you were successful as a blogger because you were driving traffic elsewhere. That's gone; maybe someone will build it, but maybe they won't.

As for the big purple squeeze of love, see also delicious, mangled by carpetbaggers after being abandoned by Yahoo. I have to wonder whether there's a web 3.0, the clipped corner remake of web 2.0 properties that were acquired and then sunsetted.

Autocomplete, letter A; as seen, somewhat edited, in my browser tonight

Note to self: if you are going to name a new site, think about when you autocomplete it what sites you're competing with for autocomplete space. 

AA - Ann Arbor District Library

AB - about:memory, Chrome diagnostic

AC - Accuweather for Ann Arbor 48104, city weather forecast

AD - Advertising Age, trade publication for the advertising industry

AE - (not used)

AF - (not used)

AG - (not used)

AH - (not used)

AI - Ailanthus altissima (Tree of heaven), in particular the Damn Arbor review of this invasive species 

AJ - Atlanta Journal-Constitution, a newspaper in Atlanta, Georgia

AK - (not used)

AL - (fails to load)

AM - AM 1610, The Station, from Hamtramck Michigan.

AN - Ann Arbor Chronicle, online newspaper that focuses on civic affairs and local government coverage for Ann Arbor, Michigan.

AO - AOL.COM

AP - Steps application on Facebook, which loads via apps.facebook.com. 

AQ - (not used)

AR - Arborwiki, a local encyclopedia of information about the Ann Arbor area that anyone can edit.

AS - ASK.com, a search engine

AT - ATT.NET, which redirects somehow to My Yahoo. The dream of the 90s is alive.

AU - (not used)

AV - AviationWeather.GOV - National Weather Service consolidated weather reporting for aviation use.

AW - Google search for AWK, a programming language that dates from Version 7 Unix (1977)

AX - (not used)

AY - (not used)

AZ - (not used) 

New design

Mostly, the design is new because I was ready to scrap a bland custom design and go with something from the Typepad gallery.  I also like this one in part because it has a nice wide area for pictures in blog posts.

I had to hit "shift-refresh" to force everything to load.

There's something effortlessly fun in doing a redesign just by picking a selection from a menu, and suddenly having everything look new.  How few things work that way!

It was especially nice that when I picked a photo to use for the theme, that the system suggested an appropriate accompanying color palette.

Also noted, 5/22/2010: as author, my top navigation bar has an "edit post" button, so I can change things well after the fact.  This is promising.  I do notice that there are no archives links anymore and no easy way for anyone to notice that I've been plugging away at this for years, but the "search" box is very welcomed, and should unearth some things that would have otherwise been buried.

3/19/11: The design I had picked was "Chroma". I've swapped that out for "Journal Black", which served to restore the category and archive links. --Ed

4/24/13: Now using the "Clean" theme, with radically simplified sidebars.

Related articles

Blogging, memory and forgetting

storing something somewhere you can remember it

retrenching on web 2.0 - controlling your presence by turning off abandoned services

If there's one thing that's characteristic of Web 2.0, it's that a lot of people made it very easy to create free accounts that let you do some very small, specialized amount of publishing on their site.  From recipes to wish lists to book lists to diaries to reviews, almost anything you can possibly think of writing had a home "free" on the net, as long as you were willing to have someone else's advertising there sponsoring it.

With the retrenching in Web 2.0 properties - either sites getting bought up and closed down, abandoned by their new owners, orphaned, or mutated beyond recognition in the search for profit, it's time to start looking closely at what kinds of remnant "internet presence" I have lurking on the internets and see about turning off anything that's no longer relevant or up to date.  Here's a list of sorts of long-forgotten enthusiasms that should be mercifully excised or republished in some current format.

Yikes!  There's a lot of debris here. 

OPML blog - Dave Winer's software for publishing; I used it about a dozen times between 2006 and 2008.  The sidebar is full of links to blogs that I don't read, helping boost their pagerank.  If I can get back to it, I'll nix it.

superpatron.com - Once upon a time I was going to move my Typepad blog for superpatron over to Wordpress; that project stopped before it really got started, leaving a half baked page up.  Redo that so that it redirects properly.

Vacuum #8 - The last iteration of my Blogger based blog from 2003, after which I switched over to Typepad.  It's a relic, and I think I'l keep it as a relic, in part because I used to have a functioning design that had a nice ever-growing ad hoc sidebar full of marginalia and none of the Typepad designs really support that well.

Edward Vielmetti - Socialtext Customer Exchange - yikes, a wiki.  It should get a copy of the standard bio.

Flickr - Not many photos uploaded since my camera died a technicolor death.  My bio is about two years out of date, and it doesn't have the current picture.  I'm increasingly irritated by Flickr's cutesiepie error messages, which don't line up with their unpleasant treatment of key staff.  When my Pro account runs out I probably won't renew it, which means I need to unpublish about 130 photos.

Jaiku - Like a bad penny, it follows you around, collecting scraps of things that you once fed it and keeping them for all eternity.  There was once a time when I used Jaiku when Twitter was down, but it's been so thoroughly orphaned by Google that it's hardly worth the bother.   

Plazes - Like a bad pfennig, it follows you around, collecting scraps of information about where you were and remembering them for all time.  I used to use Plazes, but enough creepy encounters from people who were unsubtle about the uses of always-on geolocation soured me on the use of any remote location tracking software ever again.

Scribd - Hey, once upon a time I created an account!  I hope they have password recovery.

Amazon - This bio is a few years old.  Do I really want to share my wish list - which no one has ever bought me anything from - with the world?  After all I don't really use it as a wish list, just a way to bookmark things within Amazon.

Wikipedia - Eeek!  A wiki.   Not very useful or helpful as a profile page, perhaps that's just as well.

ACM Portal - Only one publication; gotta get on that.

Pandora - I haven't used this in how many years? 

Meetup - Party like it's 2002.

Xing - If I'm going to really use this, then it needs a complete overhaul - what's there is also woefully out of date.  I'd be better off starting over.

a2skatepark - eek!  Ning, I think, gets the credit or blame for this skeletal self-description.

Plaxo - eek!  Needs work.

plodt - huh?  I never remember signing up for this, but it seems to be stalking me.

VegGuide - I apparently used this once back in 2004, and it hasn't forgetten about me yet.  Yet.

43 Things - For a brief shining moment in 2004 this was interesting; now it's just more crap on the net beyond my direct control.  I hope they have password recovery.

Yahoo 360 - I had hoped that they had killed this off, but I guess they didn't; the sleeping baby on the front page is now almost 4 years old.  (cute kid, huh?)

ListPhile - I hope they have password recovery; I don't remember ever signing up for it, but there I am larger than life.  Huh?

Vox - I'm averaging about a post every six months.

There's more, I'm sure, and I'm not even counting all of the aggregators and refeeders and etc etc that replicated stuff I wrote without me asking to or having any control over the presentation.

Widget Wednesday - Socialtext OpenSocial hackathon, November 12, 2008

Socialtext is organizing Widget Wednesday today.

PALO ALTO, CA - November 5, 2008 - Socialtext, the Enterprise Social Software leader, today announced it will host Widget Wednesday on November 12, a distributed hackathon for widgets and mashups. Socialtext partners Box.net, Denodo, Newsgator, Meebo, Six Apart, Slideshare and others will participate in the hackathon to develop OpenSocial standard-based widgets and mashups on the Socialtext platform. Socialtext also announced that it will release all of the widgets it develops under an Open Source license -- to help partners, customers and developer community members create mashups leveraging Socialtext's best-in-class REST API.

OpenSocial is the standard propogated by Google for creating widgets:  There is an OpenSocial Foundation, and they are celebrating a 1st anniversary in (where else) San Francisco on November 13. A highlight of the event is the Global Container Crawl where you test your install on as many systems as you can do in one afternoon.

There are hundreds of millions of people using social networking sites that support OpenSocial apps. To take advantage of the distribution that OpenSocial provides, you'll want your app to run on more than just one site. There are a few things to consider when writing or maintaining an app that works across containers. This page will give you an overview of these issues, and you can find tips for reaching users on specific containers by following the links in the Containers section.

The OpenSocial community wiki has way more details.

The Little Brown Jug, and other Michigan - Minnesota connections

Since it's game day (Michigan at Minnesota) I thought I'd link to at least a few connections between Michigan and Minnesota.

1.  The Little Brown Jug is the trophy that goes back and forth between the two schools depending on which one wins the game.  It was originally purchased as a water jug for the Michigan team on an away game to Minnesota in 1903, it was left behind after the game (a 6-6 tie).  The Daily Gopher has an excellent roundup of the story (warning: Minnesota point of view) quoting from several Michigan football blogs.

Days later when Yost sent a letter requesting his water jug back, Cook replied saying "We have your little brown jug; if you want it, you’ll have to win it."

More on the Little Brown Jug, and on the Brown Jug (a South University campus bar and restaurant) at Arborwiki.

2.  Michigan shares a border with Minnesota along Lake Superior, and there is public transportation by ferry from Houghton, Michigan to Grand Portage, Minnesota via a stop at the Isle Royale National Park.  

Minnesota to Isle Royale ferry information:

Grand Portage Isle Royale Transportation Line began as a way for the Sivertson family of Lake Superior's north shore to transport their catch of freshwater fish from Isle Royale to Minnesota. It gradually expanded to include passenger service.

Michigan to Isle Royale ferry information:

Unlike other Isle Royale ferry services, the Ranger III was designed by the National Park Service specifically to bridge the gap between the mainland and the Island. It was built to withstand the Lake’s fierce storms and its ice breaking capabilities assist in opening the park in the spring. An experienced nine-member crew take great pride in not only maintaining the vessel, but in serving its passengers. The Ranger III’s Captain, Bill Hanrahan, a graduate of the Great Lakes Maritime Academy, has over 30 years of experience; the crews combined experience is just under 200 years.

3.  Ann Arbor, MI was the site of GopherCon '92, a 50 person workshop for developers and users of the Gopher information system, a pioneering Internet publishing system that was for some brief furry moment the hot new thing on MTV and bigger than that super-complicated World Wide Web hypertext thing.  Prentiss Riddle wrote up the trip report for it, and here's his writeup on my presentation at the time:

RELATED TECHNOLOGIES: Ed Vielmetti of CICnet gave a talk on "what we would be gathering to discuss if UMinn had never developed Gopher", meaning primarily World-Wide Web (WWW). WWW was developed for the high-energy physics community and serves as a model of what Gopher could do if a discipline-oriented virtual community invested in it
heavily. WWW is based on SGML (Standard General Markup Language"), an ISO standard for marking up text which WWW uses to implement hypertext. SGML is a bear and it is a significant investment of effort to properly add a document to WWW, but the result is quite powerful (for instance, WWW handles footnotes in hypertext).

The Internet periodically notices that Gopher is not dead; here are a survey of the current Gopher infrastructure from Jason Kottke, some April development from George Hotelling, and a manifesto from Bjorn Karger.