Hacker News new | comments | show | ask | jobs | submit login

Interestingly one of the things that I found most amazing when I was working for Google was a nearly total inability to grasp the concept of 'deadline.' For so many years the company just shipped it by committing it to the release branch and having the code deploy over the course of a small number of weeks to the 'fleet'.

Sure there were 'processes', like "Canary it in some cluster and watch the results for a few weeks before turning it loose on the world." but being completely vertically integrated is a unique sort of situation.

Android is much like a more traditional vendor/customer model where they are at a place in the pipeline that has to co-ordinate with other things that are in the pipeline. Except when those 'other things' are hardware it gets really strange.

I was present in meetings in which all the folks at the meeting would nod their heads at some release date which was totally and completely imaginary. It existed so that dereferencing the term 'release date' didn't return 'undefined' :-). I remember my first project there I asked SRE (part of operations) for a date when they would be done with their part. Got that date, operated my part of the project against it, the date arrived and nothing. Two days before the date I had re-confirmed it, and on the date was told, "Well its not quite ready, probably in a couple more weeks." Really? One of the reasons people do 'starter projects' there was to learn this lesson.

The 'tablet' version of Android has been a planning disaster from the moment it was conceived as far as I can tell. Why isn't ChromeOS (basically Android kernel + browser UI) not the tablet version of Android? Android/Tablet vs Android/Phone? AFAICT its because Apple pulled off making iOS work between phones and tablets, and they only pulled that off because they had the iPod Touch for YEARS before they shipped the iPad and kept asking 'what is a non-phone experience like?'

I have no inside knowledge of how the whole 3.x thing went down but if it was like other projects there, rather than get some clear thinkers together in a room and put together a plan and then manage against the risks of that plan, two or maybe three 2 or 3 person 'teams' (friends likely, in their 20% time) probably started working on their vision of how it might work as a tablet, only to be out maneuvered at a review (or not) to have several competing versions going forward which then someone checked in first and everyone else was stuck following that version until someone else checked in something better. Reasoning? Planning? Discussion amongst stakeholders? Probably not so much.

So I would certainly place the underwhelming nature of Android tablets completely at Google's doorstep. Its not like the hardware is less capable than the iPads, in some cases its even more capable.

Its not in Googles DNA (at the moment) to execute against a customer/supplier type of arrangement. Fortunately I think there are some really smart people at Google who understand that, I hope they can prevail against the inevitable corporate anti-bodies that are now doubt even now trying to marginalize them.

I do believe that Android will win here but for the same reason that lava coming out of a volcano 'wins.'




Why isn't ChromeOS (basically Android kernel + browser UI) not the tablet version of Android?

Because native apps are sometimes nice. A tablet that can't play Angry Birds is a hard sell.

Interesting comments about Google's development practices. My impression is that Honeycomb's problems stem from the fact that there was a hard deadline: they wanted to get the Xoom out before the iPad 2. If that's not how Google normally operates, that would explain a lot.


Google is also not really a hardware company. Their primary focus is software.

Competing competently in the tablet market is really stretching Google's strengths as a company. I think they can do it eventually, but it's going to be a long learning process.


My impression is that Honeycomb's problems stem from the fact that there was a hard deadline: they wanted to get the Xoom out before the iPad 2. If that's not how Google normally operates, that would explain a lot.

This.

Also, I've heard that Google developers didn't know the deadline until after the Andy Rubin demo (which was a few month before launch)




Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact

Search: