Reaction to an Editorial I Didn’t Like

I got a link a few days ago to this editorial by Mary Jo Foley on a private list. As I was writing my reply I decided to make it a blog post. Then I decided to cool down a few days. Then I decided to post it. It’s not profound, it’s just my editorial response to an editorial that pissed me off.

In the IT community we become so accustomed to hearing well thought out opinions from people that are astoundingly technically competent like Scott Hanselman, Mark Minasi or Richard Campbell (randomly picking an awesome dev, IT Pro and all-round devops guy) that we can fail to take background into account when reading a professional “industry watcher” journalist. When Mary Jo talks about the implications that Microsoft’s bottom line is primarily based on enterprise income such as she does here, please read. But when she talks about what devs should be worried about for the future, please be sure you are (at least also) reading things written by experts in development. And that your boss is (at least also) reading things written by experts in development.

The problem is that experts in development don’t want jobs writing scheduled editorials covering lightweight version of heavyweight topics. I know, because I recently turned down such an offer.

Things that are correct in this editorial

Microsoft will talk about its future strategy at Build – it is the declared purpose of the event. Microsoft will pay attention to what customers say after the event.

Devs were nervous after Build 2011. And every other day of every other year of every other decade (at least dev leads, managers and any devs paying attention).

The biggest thing this editorial got wrong

Microsoft has never been more transparent regarding actual products, and I’ve been watching a really long time.

Mary Jo says “The Windows transparency debate has waged since Microsoft got burned by sharing too much information too early about Longhorn/Vista.” With Longhorn, they showed a blurry vision of something they proved unable to build, and that in retrospect didn’t even make sense. Maybe it’s a debate, but it’s pretty clear that it’s a bad idea to blind people with glitter covered papier-mâché balloon fantasies. It was a bad idea for everyone, particularly, folks like me. Ouch.

But while the Windows division tightly controls when they raise the curtain, they are not raising the curtain with less transparency on real products. They continue to wait until they think they know what the product will look like, what their overall strategy will look like, and what their marketing look like. They continue to schedule Build events and invite journalists.

If Soma or ScottGu says “hey, we might make this” and later waves their wand around and says “naw, forget that, look over here, this other thing is so much way cooler,” devs get excited. If the Windows folks do this, the stock market and enterprise customers will not be happy.

As an aside: The Windows division has also been less transparent to their deeply NDA friends than the developer tools division. This is very bad for them because it slows their progress by not getting super-early feedback. Maybe they can learn from the success of the developer division in embracing the community.

The next biggest thing this editorial got wrong

As I talked about here, Microsoft is completing a critical revamp of what you know as the “.NET Framework.” This, along with the PCL allows the technical foundation for cross-platform code and isolation of code so that it can actually be sensible across platforms.

They have also quietly expanded the very technical area of creating native code from IL (I think we are now at seven strategies) in a way you haven’t needed to worry about. Did you even know that one of the things the Windows Store does for you is prepare your app for the particular phone it’s downloading to? Phones don’t like JIT strategies due to power consumption – you never had to think about that because the core teams already solved it for you.

Microsoft is also in the process of rewriting the C# and Visual Basic compilers, combining a visionary strategy with a complete commitment to existing customers. It will be an “open” compiler, but the commitment is that it won’t hurt you, in either backward compatibility or significant design time performance issues. They are making that commitment as they move from hand optimized C++ code to managed C# and VB code. Aren’t you impressed?

They’re also continuing work on the technologies that support core platforms – including important things like support for platform production diagnostics.

So it sounds bizarre to me for Mary Jo to say “Microsoft needs to insure its developer community that it isn’t going to leave them behind as it hones its platform strategy.” You can’t live on a fifteen year old technology. It’s never happened. They rebuilt the .NET platform strategy underneath you and you (probably) didn’t even notice. Apparently, Mary Jo didn’t. There are definitely problems outside the core in libraries. But Microsoft has made tremendous investments to ensure that .NET is ready for platform change, which is what she was talking about.

I am absolutely not here to pick nits, but if this is actually a typo for intending to say they need to “assure” you it isn’t going to leave you behind, then I agree. They are doing the work, but not talking enough about it.

Other things overlooked

Whether the eight dozen Windows Phone devices you and I are currently carrying continue to run Windows N+1 is not an interesting question, and it certainly isn’t a developer question. Can Microsoft create a phone, any phone! that sells enough to make developing for Windows Phone a compelling story? Do the eight dozen of us carrying Windows Phones actually care whether we have to buy a new phone?

XBox? The game market changed. Tis done. No one understands the new market. Outselling PlayStation and managing Steam competition is just the beginning. Should XBox be your new Roku, and if so should it block Amazon (no Alpha House?). Is home security via Kinect, Mazola based couples’ Twister, or Skyping with Grandma the killer app? Or maybe XBox should be a refrigerator/microwave/coffee make combo with an auto shopping feature. I’d like that.

By the way, you’re day job is probably in the Enterprise. Which means you care more whether Microsoft can make BYOD the gold standard.

David Sobeski’s article

It’s awesome that Mary Jo helped this interesting historical piece get more attention. But in a piece that was 2185 words long, the first technology of the last decade appears at word 1178. It’s an article about the 1990’s and 2000’s.

The same breaking change problems of the 1990’s and 2000’s have absolutely extended on the phone platform. Let’s be very clear here. The market for Windows Phone is not developers. The market for Windows Phone is consumers (including business consumers). I am truly sorry Microsoft has screwed phone devs with every version of the phone. I hope that’s over, but it’s not the most important issue. It’s not the most important issue until I see someone, anyone, at Starbucks pulling out a Windows Phone (outside the Seattle metropolitan area) and my son isn’t horrified at the thought of having one. The game is getting the retail hardware and the provider relationships right. It’s the only game.

There is a problem of trust

Of course there is a fracking issue of trust. But let me make that a different, and possibly much more interesting post.

Leave a Reply

Your email address will not be published. Required fields are marked *