Slashdot Mirror
Technology Review Profiles Miguel de Icaza
prostoalex writes "Technology Review has a feature story on Miguel de Icaza, currently Novell VP of Product Technology, but more known as the leader of Gnome and Mono projects. Miguel is the man Don Box would like to see joining Microsoft for his "amazing amount of raw energy". If you read through the Technology review article, you will see that de Icaza was actually turned down by Microsoft at some point."
Is Miguel de Icaza as bloated as his creations?
I met Miguel, like, back in '98 at a conference in Mexico. Yes, Linux existed there back then!
We chatted and I quickly found he was more than just a Rob Malda or Rusty Foster, guys who talk the talk and get all the fame but can't back it up when it comes to lines of code per hour counts.
Miguel simply AMAZED me with his knowledge and skill. He ever opened up a digital projector and messed with the PROM or jumpers or something and fixed it within 20 minutes, just in time for his talk.
de Icaza is nothing short of amazing. I DO however question his judgement to kind of jump into the MS camp with MONO/.NET emulation, but I know that since he's smarter than me he must be doing the right thing.
If you liked my post,
I'd like to thank Miguel for his contributions. I'm a gnome user, and it is quite nice. What I don't get though, is why he seems absolutely fascinated with the boys in redmond. He reimplements Outlook, and now he's reimplimenting their reimplimentation of Java. Why not get behind an OSS implementation of the original ala kaffe or gcj, or push the OSS own Parrot?
I think the most interesting part of the article was near the beginning, where it described him as being both idealistic and pragmatic. That's exactly the kind of person we need promoting Free software.
RMS was both at the start of his career - and, interestingly, he started fading out when he seemed to have lost the pragmatism (GNU/Linux, Hurd, etc.). Hopefully Miguel will avoid making a similar mistake.
To me, at least, it seems like he's got the world's best job: get paid to produce Free software. Not a bad gig.
-Erwos
Plausible conjecture should not be misrepresented as proof positive.
"A lot of people denigrate Miguel as being a "Microsoft fan."
Somehow a phrase comes in from the past. "Chasing Tailights" comes to mind. The other is "Bandwagoning".
"That's not fair. What he is, is a realist. The fact is that as long as Microsoft has a vast majority of the desktops out there, any competing system has a choice: between creating their own 31337 world where only the initiated may play, or instead creating systems that work and play well with others. By paying close attention to what system and paradigms users are used to - that is to say, that Microsoft ships - Miguel helps furhter the rapid adoption of Linux as a viable Windows alternative. "
The bad thing for your argument is Apple Computers. Apple computers sets those paradigims you're so proud of. Why isn't Miguel following them? Apple's play better with Windows and Linux. Why isn't Miguel following Apple? Apple's world is no more "31337 " than the PC, for the entrance fee's the same for both. Why isn't Miguel using an Mac? So what argument's left? Money? Combine Apple's paradigm's with linux's low cost, on cheap PC's and there's no reason left for your "realist" not to follow Apple?
I was at the Burton Group conference where Miguel de Icaza, John Montgomery, and Graham Hamilton participated in a forum discussion.
I mentioned how prolific scripting languages had become, that some very large and revenue-generating systems were built on scripting languages. I asked given the industry-wide move toward virtual machines, what each of their products would be doing to facilitate scripting languages targeting the VMs.
John Montgomery admitted that the CLR did not really handle dynamically typed scripting langauges very well. Graham Hamilton did not say the same thing about the JVM, but did mention they were working on getting the JVM into better shape to be able to allow dynamically typed scripting languages more ease of integration.
absolutely agree on that point. If that is the primary concern for someone, then they should use perl, parrot or python. The comparison is apples to oranges, but the type of services is not that different. Both currently support webservices, a framework to build dynamic pages, database connections, templating, xml, and all the other fancy doodads. I'm definitely biased, but I have no problems with too many options in the Java OSS stack. It's not really one stack per say, more like a ton of stacks that are very similar and yet different.
Unlike some developers, I don't expect someone to hand feed me and I expect that I will have to take time to study and understand what each approach solves and how it fails. What I don't like is that for most of the work I have to do, .NET currently doesn't provide the features I need. In fact, the current project I am on, 80-90% of the code has been building new functionality missing from the .NET stack. Those features are availabe in the Java OSS stack and considerably more robust and mature. Even though I've tried to educate the developers at work and suggest they port apache stuff to .NET they stick to the MS stuff. The end result is the stuff doesn't work. Now obviously, that is not Microsoft's fault and is strictly the result of incompetent developers.
From my own experience, the quality of "real senior developers" in the Java OSS world is an order of magnitude better than the typical .NET senior developer. But that's my experience and isn't necessarily representative of reality. Also, of the senior java developers I've know, a large percentage of them have written compilers and have a good understanding of low level details. In comparison, of the 30 or so .NET developers I work with currently, none of that have that knowledge. If I consider the microsoft developers I've worked with or have known first hand the last 5 years, none of them had the expertise.
It doesn't matter how much code Miguel writes. What matters is if it the right code. It probably is the right code for Microsoft, because imitation is still the sincerest form of flattery. But I am still not convinced that it's the right code for Unix. Microsoft may have the vast majority of desktops, but Mono will do nothing about that, because those desktops will still be Windows, and they'll still be running .NET code.
Don't blame me, I didn't vote for either of them!
What an odd coincidence - we seem to be using our computers at the same time, imagine that amongst slashdot readers :)
.ASP thingies, and they're always terribly written (always VBScript btw.) - I do my best to sway my customers into using PHP solutions, because that way I know the code I do for them will work on a variety of OSes and Webservers.
.NET, because no matter how elegant C# may be, and no matter how elegant the layers beneath the language may be - it's right now a surefire path to vendor-lockin.
.NET is out of the question for me, and I'd rather drive a cab (I kid you not - I am getting the license right now) than do MS/Oracle/RandomBigCorp/etc... only solutions. It simply is no fun to know that I in part work to help a big vendor maintain it's grip on computing.
.NET ECMA spec, and if so what makes you believe that ?
I am a total vendorlockin-phobe, I do "small-time" webapplications - and people come to me with their
I really shudder at the mere thought of
As long as MONO is still infant (or is it adolescent by now ?)
There is a lot of speculations around about whether MONO is playing a realistic game. Will M$ just strangle MONO if you get too close etc... and I have very little knowledge to help me judge on that.
Is there any where I can read the MONO viewpoint on this issue. I would love to see a FAQ type document addressing these concerns.
For instance: I know not enough to understand the implications of the ECMA thingie, but I can't help thinking that Javascript has an ECMA spec (ECMA script I believe) and that MS does not adhere to it fully.
Does the MONO community believe that MS will stick to the
So in short:
Did the MONO community consider 'worst-case-MS-behaviour' and the following worst-case-scenarios ? And if you did, is there some where I can read about that ?
Check out my PHP Url Validator
In the crowded world of Unixesque/Linux users, developers etc this is the one guy who gets it.
.NET to linux says it all. There are reason why Microsoft advances so far and no...its not simply because of windows dominance on the desktop or marketing. MS understand where modern computing, programming needs to go and so does Miguel.
His writing on Mono and why he want to port
Well consider it this way. If Microsoft decides to break
You should maybe look into IBM's SWT. I'm using Azureus, a Java bittorrent client using this toolkit, and it integrates fairly well with my Gnome desktop. It even puts an applet in the notification area.
The flagship SWT app is of course the Eclipse IDE.
I also hear Java 1.4.2 includes a GTK look & feel for Swing. Hopefully the Jedit texteditor I use for coding will be updated to support this.
Mono can be blown up at any time simply with a patch. Is this confusing any luminaries out there? Is this too deep?
.NET apps, too. Great idea. Also noone forces you to connect to a MS server, you can also write stand-alone programs in Mono, and they can talk to whatever backend you like. So what's your point again ?
Yeah, by that blowing up truckloads of existing
If anyone belives mono is evil, downfall of gnome etc etc i recommend you read the article. Listen to what Miguel says.
.net? According to MS that is a yes. What if all that windows code can run on linux to, without problems, seemlessly. Mono is the key. A compablilty layer, just like wine expect better, to allow us to run the next generation of windows applications. Its a drive at the future market itself. Trying to get a head start on microsoft with a strong development tool kit. If we wait 5 years before mono is built then we will be chassing there tails yet again.
We are talking long term. In 10 years will 90% of windows software be written in
I don't think someone who gives the impression he'd rather be a M$ developer than a *NIX developer should be considered a "leader" in the community. I won't deny that he's done some good for the community. Yet, I don't consider MONO one of them. I think it opens the door to possible problems down the road with M$. I feel the best way to approach the war with M$ is to open standards. Not concede that M$ has created them already and conform to them if for no other reason than to avoid litigation.