Regarding persistence

"Those months schooled me in patience: now I could propound a complete theory of patience in several ponderous volumes. The gist of it would be that after every failure it is necessary to examine yourself and ask, 'Look here, perhaps you are wrong?' If you are wrong, get off the rostrum and shut up. But if you are right, do not believe your eyes: that defeat is not a defeat. 'No' is no answer. Wait an hour and start right over again from the beginning." - Vladimir (Ze'ev) Jabotinsky, writing about his attempts to organize a Jewish Legion to fight the Ottoman Turks alongside the British in Palestine during WWI (quoted in pp 169-170, "Lone Wolf" by Shmuel Katz, publisher: Barricade Books, 1996).

Google Buzz vs GMail

SortiPreneur makes the case that Google knows the concept of "open" and they're building Buzz as a platform. I originally didn't like the fact that they didn't integrate with Twitter directly (which I'm sure they will do, I agree). So, no argument there. However, regarding this:

For me, the bottom line is, for a gmail user, the amount of social insight residing in gmail far exceeds facebook. This may not be true for users who split up email usage for personal and professional communications.

Yes, precisely - this is the scary part! I don't split up my email accounts, or rather, I may have different email addresses, but the interface I use is one gmail account. Most people are not going to be comfortable with email being mixed in with Twitter/Facebook. I totally agree that Buzz is going to be a "platform". With the new promises of a separate googlebuzz app, I might use that as an "outward facing" communications platform. I like the platform aspect (to unify Twitter/FBook comm for example), but I feel uncomfortable with email & chat being mixed in automatically and treated as equal communications channels on this platform.

For me, email/chat should be separate, and not have the risk of mixups (e.g. facebook contacts getting info into my email life such as who I email). Privacy controls are not enough - they can, and do, fail:

• Bugs : Already the fact that Google Reader users could see items even if they were blocked from Buzz
• Changes to privacy policies in the future. Remember the recent Facebook fiasco, where the "redesigned" privacy controls by default changed all of your information to public - and also even now some information remains unhideable.
• The user can easily make a mistake - many users complained about Buzz allowing people who email them frequently but whom they definitely do NOT like like their ex-husbands or landlords, seeing their updates.

This problem cannot be solved with more careful quality control or better user design. It's a fundamental problem of the same platform being responsible for your private email and public internet life. People have said that "privacy is dead" but I don't think they meant this in the sense that they expect the world to read their private emails. So even in the modern world, there is still some expectation of privacy. Splitting up the platform at least at the user level, if not at the company level (in other words, still having Google as a company to run both email/chat and public communications services) would allow the company to profit from both sides but with some control over privacy. Google for example could do it like ads in GMail (maybe if you email a lot of about cars, you might get car-related ads on your buzz listings) - this still compromises some privacy but they could put in controls such as no indecent ads and no sensitive subjects would be shown. This isn't the same thing as showing your email contacts to the world.

For myself, I'd like to have 2 platforms, one for public-facing and one for email/chat. I realize of course that 1). This isn't good for the companies making money from it. and 2). this isn't necessarily the distinction that everyone might want. They might prefer work vs social (private & public together) division, or no divisions. But it just happens to be what I want.

Poor Pandora, and some Privacy Surprises

So I downloaded and installed (btw, shouldn't we already have a combined word for "downloaded and installed", like "dinstalled" or downstalled or something...) PandoraJam so for $15 or so, not only can I let my friend(s) see what I'm listening to at the moment (on my IM status), but also deprive Pandora of the last remaining vestige of hope of ad revenue.  Poor Pandora... 

When I installed PandoraJam, it instantly connected to Pandora, not asking me for my username and password.  This reminded me of something that I found out a while ago, but I think most people are not aware of: a Flash player keeps its own cookie-like local information (up to 100kb by default) where it can store things like credentials - or track your surfing habits.  When you clean up your "private" data like cache & cookies on Firefox or Safari, it does not clear that cache.  To do this, you'll need to go to the Flash Settings Manager page to clear the information separately from your browser's mechanisms.  For those people who religiously clear cookies (or set them to be cleared on browser exit), you might be surprised how much persistent data is there.  This data can (potentially) be used to track where you surf, by using flash-based ads on various sites.

Google Spreadsheets Rock! Google Sites is the Web2.0 equivalent of a Pet Rock.

Just saw that there is an API for getting data in and out of Google Spreadsheets.  This is interesting: combined with the fact that I can get feeds of what's changed, collaborate with others, this makes Google Spreadsheets an awesome platform.

This is more than I can say for the hopelessly underpowered Google Sites (JotSpot).  I'm using it for our intranet, and the biggest issues (this has been covered before, but here are my top three biggest problems with it):

1). Unlike Spreadsheets (which can be collaborated on real time), Google Sites locks everyone out when you're editing.  And it takes a while to release the lock, even after you've saved the page.

2). Unlike Spreadsheets (which has a great API to get data out), Google Sites has NO way of extracting data.  None, nada, niente.  If I want to copy-and-paste a LIST page from Google Sites, I can't even do that, because it's all ajax-y and you can't copy-and-paste.

3). Unlike a Wiki, you can't create a new page by just typing the name.  You have to (a) Save the page you're working on.  (b) Create a new page (c) put some content in it, (d) go back and find the original page (e) link to the new page, by finding it in a list (which could be huge) of existing pages.   What a pain.

ICEfaces + Eclipse + Seam

So, I got tired of looking up ICEfaces tags in the documentation, and downloaded the ICEfaces plugin for Eclipse from here.  Then I tried to use the code completion feature.  No dice!  According to this bug, it's broken, and isn't likely to be fixed soon.  Ouch.   And I thought it was going to be smooth sailing.  Apparently this has been a problem since at least 2006.  The best thing I could do is to rename the "xhtml" files to "jspx".  Hope that Seam still works after this :(

Go RonR! wait, go Seam! wait, go WebWork!.... argh...

In my new venture, GoalSpring, we're trying to select a framework to develop in.  Every time I talk to Dave Astels, I get the urge to develop everything in Ruby.  Use RSpec to do real BDD (Behaviour-driven Development), develop everything the "right" way, etc.  Then I realize that besides one two-hour seminar, I've never written a line of Ruby code, and I'm not sure our investors will fund my learning curve.  So yesterday, I've decided to go with Seam.  I dutifully went through the examples, and hey, besides no Rails migration (my favourite part of Rails that I've found so far in my limited experience), Seam seems (no pun intended), pretty equivalent.  And of course I get all the familiar Eclipse refactoring support that I expect.  None of this emacs-like "TextMate" development that all those Ruby junkies seem to enjoy so much.   When I want to write a method, I actually LIKE the method completion and all the help that Eclipse gives me (or IntelliJ, but you get the point).  

And I thought that Green-field development was easy!!

Then I looked through some of the other frameworks that are around these days: WebWork/Struts2 (can they at least drop the unfortunate "Struts" name at least?), Tapestry (I've used it a long time ago but now it seems to be better and cooler - awesome).

Looks like I'll stick Seam & the "no one got fired for choosing J2EE" way of selecting software :)   (well, ok probably people did get fired - but I digress).

Happy International Workers' Day!

http://www.foxnews.com/photoessay/0,4644,1695,00.html