Sunday, 29 September 2013

The "Rockstar" programmer

Scott blogged about the Rockstar programmer. Maybe a coincidence, but yesterday I read this other article in the same topic. Then I found this from Joel. They all accept the existence of Rockstars. However while Scott and Shanley emphasize the importance of "Rockstar Team" over "Rockstar Programmer", Joel says that you have to build a Team of Rockstars. The context is the difference here and Joel is the only one aware of that. As he rightly pointed out, a successful software company has no choice but to build a team of Rockstars, while in-house development rarely requires or can afford that. I would go further. Because mediorce/average programmers would struggle to work with rockstar's code, it is in the best interest of the in-house development to avoid rockstars. They won't enjoy the environment and they will leave. Their replaceability (however inappropriate this word sounds here) is limited and very expensive.
Pairing "step back and think" devs with "crank a lot of pretty good code out" devs is a recipe for a good team. - Scott
"Crank a lot of pretty good code out" doesn't make a Rockstar in my dictionary. Rockstars don't need somebody else to "step back and think" because they do that when necessary.
Calling out rockstars demotivates the team. - Scott
If you call yourself a Rockstar, you are not one. - Anon (I can't find where I read this.)
Besides, if your team is like Joel's then this would only motivate the others to become even better.

And a funny side-note: Scott works for Microsoft. Microsoft is a software company. Joel uses Microsoft in his examples. And earlier he also justifies the success of Wal*Mart. Is Microsoft the Wal*Mart of software companies according to Joel?

Tuesday, 24 September 2013

CyanogenMod and Android fragmentation

Android's OS fragmentation is an ongoing issue the Google mobile ecosystem faces. Whilst iOS7 adoption hit 35% within a day (that's including my 3-year-old iPhone 4), Android Jelly Bean stands at 45% more than one year after the release of 4.1. Some companies, like Amazon with its Kindle Fire and forked Android OS, are obviously working against Google's best interests. But what about the recently incorporated CyanogenMod? Before the incorporation I agreed that Cyanogen was helping to unify Android. But this changed last week. Sooner or later they have to make money. Some options the same article cites:

  • "Licensing software/services to OEMs" OEMs make big money from their short device life-cycle. Cyanogen works against that. Also good luck to do all the certification (second part of the article) work that holds back the software upgrades, especially for low-budget, low-profit devices.
  • "Building hardware" In some way Cyanogen's sole purpose is to make all Android devices equivalent. It will be really hard to give a reason a customer should buy their devices.
  • "Creating secure enterprise solutions" This one sounds interesting. I am not sure if they can do it without forking Android.

All in all, it is possible that Cyanogen keeps helping with the unification, but it's way more likely that in the long term in one way or another they become another "Amazon" in the Android universe.

Monday, 23 September 2013

Actually usable Windows tablets on the horizon

And no, I am not talking about the new Surface "tablets". Beginning this September Intel announced their new tablet processor family built on Bay Trail architecture. This is a great news for those who plan to buy a Windows tablet this year. Including me :) As of today there are three options:
  • Windows RT tablet. That's a joke, not an option.
  • Full Windows tablet with non-Atom processor. And with CPU-fan. And at least 800gr weight... that's not a tablet.
  • Full Windows tablet with Atom Z2760 processor. And 2GB memory. That's a joke again.
With the 3000 series Atom processors we can have a full Windows tablet with 4GB memory, no CPU fan, and ideally also in size 8" and sub-1-pound weight. Like one of these. For the 64bit versions we have to wait until Q1 2014.

About Windows 8.1 "Smart Search"

In October 2012 it caused a small "rebellion" against Canonical when they released Ubuntu 12.10 with built-in Amazon search. The problem was that users' search was sent to Amazon even when they intended to search locally. That "feature" wasn't welcome in the Linux community to say the least. Now, just one year later, Microsoft is going to release the same "feature" in Windows 8.1. The big difference is that users can disable it. However, watching the silence of the Windows community and knowing the willingness of the average Windows user to fiddle with system settings, I am sure Microsoft copied a perfect solution to monitor their customers. The rare moment I must admit: albeit unethical, this was a very clever move.
And as a side-effect, since the search engine can't be changed for this feature, we can expect a serious gain of Bing's market share; a search hit for Bing every single time a user does search on their desktop.

Wednesday, 4 September 2013

Microsoft buys Nokia, Ballmer leaves

Some say it makes no sense, except if Nokia was about to switch to Android or to bankrupt.
Cringely says it's all about money-laundering.
A long summary from Tomi Ahonen.
Some are visioning the end of Windows as standalone product.
And now everybody is like Apple. Except that they are not (see below).

I, personally, think Microsoft just made another non-decision decision. Now their mobile ecosystem is neither Apple-like nor Android-like. It's the mix of the two and that is simply wrong.

And by the way: the step down of Steve Ballmer is the greatest news of 2013, or maybe even the next decade. Vista, Longhorn, mobile, Nokia, tablet, Silverlight, and just recently he tried to ruin the Xbox business. Anything comes, the future can only be better.

UPDATE: I've found a good article describing Microsoft's mixed "strategy".