Google announced a new plugin called Google Gears that allows web applications to provide functionality even when users are offline.  Right now, the only application that is using the new plugin is my beloved Google Reader, so I downloaded the plugin and decided to try it out.

Most of you who know me well are well-aware of what a Google whore I am, so this is hard for me to say– Google Gears sucks, and it sucks hard.

So there I am, flying through my feeds at my usual clip, when a pop-up informs me that I’m no longer connected to the Internet and asks if I would like to work in the offline mode.  I switch over to a different tab, type in www.yahoo.com, and it comes up immediately, so I’m thinking that Google Gears is mistaken.  I switch back to the Reader tab, close the pop-up, and I continue reading without a hitch.

And then the pop-up happens again.  And again.  And again.  And so I disabled Google Gears, and I’m iffy on ever re-enabling it.  I dig the idea of being able to read my feeds offline, but not at the expense of my user experience when I’m reading them online.  Notifying the user that the app thinks your offline is the sort of thing that should be handled by a status icon in the upper right hand corner right up until the point that the app can no longer satisfy the user’s request– the way that true desktop apps like Outlook handle it.

Blergh.  Very disappointing.

Labels: tech, web2.0

[ 1 comments ]

there’s no way scoble’s second mix07 prediction is serious, right?

the idea is that you’re going to take something that’s slow and put it on top of something else that’s slow and somehow that will magically make twitter scale?

Labels: tech, web2.0

[ 0 comments ]

sean kicked off a thread on awful VC elevator pitches of the form “it’s like ___ but for ___ instead of ___” the other day, after we read this piece on techcrunch about scribd, which fits the template as “it’s like youtube but for documents instead of videos.”

so you can imagine the horror i felt when i woke up this morning to read that scribd might actually be succeeding. even though the source is techcrunch, the initial pageview numbers and what not look pretty healthy. i’ve even seen some stuff that was posted on scribd make it to digg’s front page.

where i’m thrown in all of this is that we already had a ‘youtube for documents’ before scribd came along. it was called ‘the world wide web’, and it’s been working pretty well for a few years now.  i’ve been trying to figure out what additional utility scribd provides beyond what the web + browser already did, but nothing comes to mind off hand– even the moderately-useful feature of converting a document into an mp3 is provided by a third-party service.  to be fair, they bring some useful (albeit pre-existing) utilities together into a single place, but at the cost of the author handing over quite a few rights to scribd to do whatever they want with your document (see section 7.2 of their terms).  it’s the intellectual property equivalent of selling your soul for a bag of cheetos.

i have a different complaint about the “youtube but for data” companies, like swivel or ibm’s many eyes.  first, they’re not like youtube in the most important way: super-frickin’ easy uploads.  swivel only accepts csv files or google spreadsheets, many eyes has a full page explaining the proper way to upload data into their system.  but this sort of constraint makes no sense when 99% of people do their data analysis in excel.  make it super-easy to upload and download the data in the eleventy zillion different excel file formats.  first and foremost, if you want to be the youtube of ___, you have to get that one right.  if nothing else, scribd has showed that people are too lazy to use the ’save as html’ option in word, so there’s no reason to expect people to use the ’save as csv’  option in excel.

second, i can’t create the kind of experience i want to give users with the tools swivel and many eyes provide.  for video, there are certain narrative structures and patterns that people are aware of and can create using simple video editing tools, but the final product is always the same- a linear sequence of images and sound, just like a movie or a TV show.  data visualization can (and i believe should) have an dynamic and interactive quality to it– like a video game, where the user is exploring a small universe in a nonlinear fashion.  the challenge– and the opportunity– for these sites is to develop the tools that allow users to create these experiences by building out and tying together basic data visualization metaphors.  the trendalyzer toolkit google acquired is a step in the right direction– the gapminder visualization is stunning and loads of fun to play with.

Labels: web2.0

[ 2 comments ]