Slashdot Mirror
Miguel de Icaza on Longhorn
An anonymous reader writes "In Miguel de Icaza's latest blog entry the Mono project leader discusses the threat Longhorn's new technologies and frameworks pose to Linux and open source. He also directs users to this recent USENET post about the goals of Mozilla, which is a very interesting read.
Originally seen on OSnews." Mmmm...Miguel smart. Seriously, good commentary - and ripe for discussion/flame wars.
Hey - if someone wants to make a goal for Mozilla, here is a good one:
Create a "drop-in" replacement for Internet Explorer. That is, it has the same layout and "feel" of the IE browser without all that monopoly crap.
I'd use Mozilla if I could shift+click and get a new browser window. But every time that I install it, I end up removing it because of little annoyances that happen from my IE habits. I can't expect to make others use it (I deploy many PCs) if I don't do it myself.
Life is the leading cause of death in America.
The "Microsoft Network" lost out to internet because W95 shipped too late. Let's do the same with Longhorn!
It is interesting that he acknowledges Mozilla's work. XUL has the potential to supply a platform that could nullify Longhown's advantage before it hits the streets.
And if you thought that was boring you obviously havn't read my Journal ;-)
Having something similar to the Microsoft platform would encourage developers to develop cross-platform. If a usable subset is developed on mono, the restriction to that subset is the price for a cross-platform application - better than a reimplementation.
What do you mean 'wasting time'?
His post is all about getting something working out of the door first. The point is defining what you need to do and how to go about doing it. Someone has to mull all of this over, privately and publicly, and Miguel's one of the ones doing this.
Good for him.
(Did I troll feed? Sorry)
I'm all for Mono, software should be cross platform, and and it would be nice to see this succeed where Java unfortunately didn't.
Microsoft's threat to the Linux community will not be raised by Longhorn. I doubt that Microsoft's newest OS will have anything drastically new that the Linux community does not already have, or that can easily be added.
Although I don't think the OSS community should be making descisions based solely on Microsoft's heading, I don't think ignoring them is the way to go either. I do think the fact that something like mono exists makes one less argument *not* to make the switch to linux. If you support .net the Linux platform can attract developers which would otherwise be coding for and on Windows only.
Just my 2ç
It's not a clone, it's an evironment that allows portable code. That is one of the points of .NET; with a written VM, the code can run on anything. Like Java, except Microsoft isn't writing the VM's for other platforms, it's down to the users, hence Miguel.
> There's no reason for me or anyone else to buy Longhorn EVER
Just as there was no reason to buy Windows XP. But still, many people did it. And new computers come with Windows XP, so there is no easy way to avoid it.
Especially when the first applications are written that only run on that version of Windows. (Either XP or Longhorn.)
I thought Sun Solaris was the standard for servers? I thought Apple the standard for schools? Face it, there is no standard. The standard software is the software which comes first. Microsoft steals standards but never sets them. The browser was not invented by Microsoft, Netscape was around long before IE. Just like Google owns the search engine and AOL owns instant msging, Microsoft is not standard. We should take the good features of OSX, BEOS, Unix and Windows and make the best OS. Forget about copying Microsoft as if Microsoft innovates. Copy Apple, at least Apple actually does invent new technologies instead of just new buzzwords.
People don't exist to serve systems, systems exist to serve people.
The 'bundled' argument may or may not be over used but it doesn't stop it from being true. Microsoft do use their market share to not just bundle apps but attempt to impose a standard. When you have 90% of the market share this is pretty powerful and uncompetitive.
g ement
The EU just gave them a financial slapped wrist over this very issue - http://www.google.co.uk/search?q=EU+Microsoft+Jud
If the OS blocked the user from installing other software that would be one thing, but they don't and you can install whatever you want to.
And just how many people do that? If you want a clue, look at the adoption of Opera, and especially Mozilla (which doesn't have the cost barrier Opera hase) against IE. Despite the fact that IE is a security-hole-ridden pile of outdated junk and Opera and Mozilla beat it hands-down on features and standards compliance, huge numbers of people still use IE. Why? Because it came with the computer and they either don't know there are alternatives, don't want to know or aren't allowed to use them because they "aren't supported".
Certainly he has boundless energy, but many people were allready pointing out that it could be the case to concentrate on getting P&P functionality with what was allready available (and hence beating MS to the market), rather than play MS at a game of catch up that you could never win (they make the rules).
And if you thought that was boring you obviously havn't read my Journal ;-)
That was quite an interesting read. While I don't totally agree with every point, the gist of the blog is right on target. I think one of the keys to Linux fighting off such threats is to get better cohesion between GNU projects, outside the Linux distro. This weekend I went to install some GNU software on my WinXP Pro laptop. I get to the download page, and ooops! I also need to install 3 other GNU projects just to get the software I want to work. Then I get to one of the other projects, and ooops! I have to install another program to get it to work. To install one app, I had to install 4 others, which meant a lot of navigation and downloading. No sweat. I am a coder; I can do this. But it did take extra time. I started wondering why these were not all packaged together, or why the installer could not simply detect they were not there and install the needed apps. This is one advantage MS has over many GNU projects and the Linux community. They are one company, and can enforce product compliance, etc. The point I am leading to is this: if the GNU community wants to beat MS in the long run they need to make sure more of their apps can easily install on MS boxes without having any knowledge of programming, IT, etc. Once you get people using this software, the switch to using this software on Linux will be much easier. Open Office is a great example of this. I know most GNU projects compile on Windows (or will with some modifications) but it has to be easier for the Windows user to get said applications.
Great ideas often receive violent opposition from mediocre minds. - Albert Einstein
Yeah.. I want 4D icons.. when I open a program, I don't want the menu icons appearing /now/ I want them appearing 5 minutes later. But i think this might already be implemented in <insert least favoured desktop environment> and I just used to think it was slow due to bloat.
If Miguel thinks that Longhorn is such a threat because it will incorporate the .NET framework, will he come out an admit the truth: that spending all that time and effort on Mono was a mistake and a waste? Trying to reinvent .NET for Unix/Linux never made any sense to me, since the components in .NET that people really want aren't available on anything but Windows. Perhaps this is a shift in his POV as a result of Ximian now being part of Novell, and they are now aiming their sights at trying to dent MS's lock on desktop and market share in the server arena. But no, he sees Mono as part of the potential answer to Schlonghorn-- don't you get it Miguel? .NET was an "embrace, extend, extinguish the competition" move to do Java one better. What makes you think that sticking with Mono will work when MS might well modify the .NET framework by the time Longhorn comes out so as to make it unusable by anything but Windows? Better to start making your own framework now instead of waiting around to see what MS will do.
Always look on the briight side of life! (whistle, whistle)
Mono will fail along with .Net, the technology is good but depending on Microsoft is bad.
.Net or it dies.
.Net .Net is a good technology, that is all it is. XUL is good too, so is QT.
Unless Mono seperates itself from Microsoft completely as a stand alone replacement technology, I don't see a use for it to even exist. Mono must be better than Microsofts
Also I wouldnt give up on Java just yet, with the embedded market picking up steam Java has a bright future, brighter than
People don't exist to serve systems, systems exist to serve people.
Despite the fact that IE is a security-hole-ridden pile of outdated junk and Opera and Mozilla beat it hands-down on features and standards compliance, huge numbers of people still use IE. Why? Because it came with the computer and they either don't know there are alternatives, don't want to know or aren't allowed to use them because they "aren't supported".
And because webmasters, especially those using Windows Media, are too stupid to embed multimedia in a way that mozilla can handle (i.e. no ActiveX, dummies). Especially big commercial sites with loadsacash budgets tend to fuck this up, whereas joe schmoe geocities sites tend to actually work (before their bandwidth limit is reached).
Most "IE-only" sites (that don't use javascript to kick you out) work perfectly in mozilla, mostly the windows(multi)media/plugin infested sites suck ass.
SCO employee? Check out the bounty
For example, why is everyone trying to get 2D vector icons when it's obvious 3D or even 4D (fourth dimension is time) icons are the way to go? Why on earth would 3D or 4D icons be the way to go? It's a simple picture that reperesents an application or idea. The simpler the better. Nobody cares if their word processor icon has phong hilights and the like.
From Miguel's blog The sandboxed execution in .NET [1] means that you can visit any web site and run local rich applications as oppposed to web applications without fearing about your data security: spyware, trojans and what have you.
That's true...if Microsoft can get it right. But as in any complex software system, there will be bugs, and considering the scope of Microsoft's deployment base, it could be disastrous. I do not think Microsoft makes worse code than anybody else, it's simply that updating their massive install base is very difficult once bugs are found. Also, the majority of Windows desktop users make poor systems administrators, there will always be bugs and crackers that exploit them. Sad, but true.
"Backups are for wimps. Real men upload their data to an FTP site and have everyone else mirror it." -- Linus Torvalds
I'm all for Mono, software should be cross platform, and and it would be nice to see this succeed where Java unfortunately didn't.
.Net/Mono, C et al will be for the overseeable future. (Dont give me the "C is portable too" crap, just today I found differences in the behaviour of strtok between platforms, not to speak of "compile everywhere").
.Net", hopefully they will interoperate more or less seamlessly, something that there is already work in progress on in more than one place.
Its always interesting to see people dismiss java as a failure out of hand with no real arguments for it. Did it fail? Depends on your point of view. Is java cross-platform? Most certainly is! And will continue to be so to a bigger extent than
Is java a failure on the client? Well, as far as circulation goes, probably, but that has three main reasons:
1. Higher learning curve, VB will always be easier to learn.
2. Old myths die hard: yes, Java was slow and java interfaces where ugly and clunky. 5 years ago! Newsflash, Java has moved forward in great leaps since the days of Java 1.1
3. Applets are mostly useless. But: Java != Applets!
Java is a great success just about everywhere else BUT on the desktop computer though, there are millions of java-enabled handsets, there are tens of thousands of java server deployments etc etc.
But.. Hopefully in the future I wont have to choose "java or
XUL does not have the potential to do this for one simple reason. Almost nobody runs Mozilla. Yeah, I know we all do, but in the real world, on the desktops of people doing their internet banking, their web based email and so on, nobody uses Mozilla, so people cannot ship a web based app using Mozilla tech (XUL). It would have to run on IE to have even a chance in hell of nullifying Avalon, and XUL simply does not, and will never, work on IE.
You're deceiving yourself if you think XUL can do it. Microsoft's new technologies WILL be out there, and they WILL succeed. If you accept that, you can be smarter about things. Let's get interoperable so we can compete - THEN we can extend into a new arena.
Miguel "gets it." The future of the web is seamless, safe perfectly integrated rapid application delivery. Imagine delivering an app via website that used native widgets and looked and felt like part of your OS, all while safely sandboxed. It's gonna happen come the Longhorn./NET heydey.
Many fanboys bitch and moan that Miguel laps up the Microsoft swill and ensures their success, but I'd argue it's the converse: Miguel knows we need to reach interoperability to have a meaningful competition in the first place. The better technology doesn't always win. Sometimes you gotta play the game via the home team's rules before the league lets you vote to change them.
although this might appeal to some users, this isn't exactly a technology that any market will jump on, especially developers. if every application requires hardware acceleration to work just to see the icons in the app, developers will be very much turned off by the idea. furthermore, linux desktops already look better than windows.
what linux needs is a technology that windows simply cannot produce without bending over backwards. and i do not mean simple UI enhancements, like tabbed browsing, or virtual desktops... because plenty of people don't like either (mostly because they're different). i mean something that makes development easier and faster. if you get the developers to switch, you'll get the users to switch in no time. problem is, developers don't see a reason to switch, when they are perfectly happy with what they have on windows (this includes incompetent develops who don't know what a compiler is, and just like to click things and write code). 3D icons won't do the trick.
BSD is for people who love UNIX. Linux is for those who hate Microsoft.
From what I've seen, most linux users are always comparing linux to windows 95 and 98...most of them having bailed out of using windows around then...and they basically are fighting against the ghost of windows past. Whereas I don't see many of these people ever saying "yes, I use winddows xp / server2003 almost constantly in an attempt to understand what I'm up against here."
slashdot, news for crazed liberal socialist zealots
People arent buying typical PCs anymore from Typical OEMs like they were back then. Microsoft had a complete lock on OEMs, the American market is sold up in terms of the Desktop computer.
New devices which run Linux now actually arent so typical in nature, devices like PDAs, Cellphones, Tivos, Video Game Consoles, Small FormFactor PCs like Shuttle and Biodeq, all come with Linux not WindowsXP.
Sun's Linux Desktop is actually catching up to Microsoft in certain countries and even surpassing them. Microsoft is King only in one specific area, the USA PC gaming desktop PC.
For laptops people don't care if it runs Windows,Linux or OSX, For media centers people don't care as long as it works well with their Tivo. For their PDA they just want it to work.
The era of Microsoft is coming to an end, Microsoft knows this, most people on the inside know this. Microsoft simply cannot sell another version of Office and another version of Windows. Most people who use Microsoft office have no reason to buy a new PC. Most people who buy new PC products who arent gamers are doing so to do new things which Longhorn currently is behind Linux at doing, like running servers, or hosting a media center, more sophisticated uses which are appearing now that people have more than one computer in the house.
People don't exist to serve systems, systems exist to serve people.
This "analysis" is poor:
Where is "Java" in that list? Java's only big problem, at this point, is the mindset that "something is wrong with it". It's really quite good, and there is a growing ecosystem of open source stuff (see SWT and friends) growing around it.
Every major computer company besides Microsoft supports it, and a Sun JVM now ships with many (most?) new Windows PCs. Even if not, a broadband JRE download is only a couple of minutes...and ~40% of U.S. households are on broadband if I remember a recent article correctly.
There is also plenty of effort going into Free/OSS JVM development, including gcj and IKVM on Mono.
Java tends to break the MS monopoly...Mono/.Nyet tends to lock it in. Which do we really want?
Galileo: "The Earth revolves around the Sun!"
Score: -1 100% Flamebait
Around 10% or so run Netscape/Mozilla, still a small amount. This amount could easily rise if AOL wanted it to, but until AOL decides to do so, Mozilla won't gain much support at least not in the USA.
In other countries however this is a different story.
If AOL were to market Netscape like they do Winamp and AIM, everyone would be using it instead of IE. We use AIM and ICQ over MSN already even though MSN comes with the damn OS.
People don't exist to serve systems, systems exist to serve people.
That USENET post goes on and on about XML formats and such and I'm not saying that's irrelevant, but XML is really more of a concern for people in specialized projects. I thought that was the whole point of XML. The browser just has to follow the standards.
I think in the browser game it's the little things like pop-up blockers and being able to manage your configs across multiple desktops are what make Firefox kick ass all over IE.
These are the things that closed source has no reason to compete on. It doesn't make anybody money to prevent ads. There's no way MS is going to compete on that front, and yet it's a huge factor for most end users.
Let's just hope that never happens... Is there anything around at the moment that ONLY runs on XP?
I help blind users with access to computers, and the evil JAWS screen reader package ($1800!) comes with limitations; you can only install it on win95, 98, ME (why?), and XP home only. No win2000 of any flavour, and no XP pro. The reasons for its restrictions are not technical though; they are built in to ensure that corporate users are charged more than personal ones...
I've started teaching one of my clients some linux skills as X can now talk... the Linux revolution is here for the blind community, as it is for the rest of us!
Real stupidity beats artificial intelligence every time.
-- Terry Pratchett, Hogfather
Freenet runs fine, and Limeware is java too if I remember right, along with lots of other p2p apps
People don't exist to serve systems, systems exist to serve people.
Also to assume Microsoft will win, is to have sold out. If you think Microsoft is going to win at everything they do, why don't you go work for them and help them.
That's the difference between being a realist and an idealist. It would be ideal if Microsoft wasn't a guarantee, but it is for now. Accept it and maybe we can do something about it.
Developing (say, mono ) to prevent platform lock-in is a hell of a lot better than trolling Slashdot and whining about how everyone else's actions are wrong.
It's a big world out there if you look out the windows - there's more to computers than the receptionists PC or using a desktop PC to replace a Playstation.
The unix way is to have configuration files in known locations with sane names (generally not one of the names of the three stooges followed by a string of numbers), and to use pipes or ports. Miguel with gnome followed the windows method of weird OLE stuff, heaps of temporary files and multiple configuration files that may as well be registry entries (gnome panel is the worst offender, including the file names mentioned above). In the end once the political stuff and odd dependancies dropped out we were left with a usable set of programs that look good - whether it is because of the design or in spite of it I'll leave to someone who knows more about the aspects of gnome that just confuse me.
Maybe I'm on a fruitbreak or something, but why not pick up GNUstep and enhance that? That way you get some semblance of source compatibility with Mac OSX Cocoa apps. Why follow Microsoft's example? It has always ended in tears in the past.
Stick Men
I really don't know what it's going to take for GNU/Linux to move out of geekdom and on to the average desktop. What I do know is that Longhorn looks like a mighty nice piece of technology that may be a winner with anyone that can afford the machinery.
I think GNU/Linux is the best thing since sliced silicon. But the other day I was reading a magazine review of a beta (or alpha or something) of Longhorn. I was deep in thought, wondering how the OSS hackers would cope with WinFS, .NET functionality and so on. I was also cynically musing about the same old security holes that'll presumably crop up, and resting assured that my choice os OS is the correct one.
A mate rings up, having read the same review. "Hey Cammo," he shouted down the phone. "The windows flap when you move 'em!!!"
Where do I order the beta?
Cogito, ergo sig.
I don't know nor care much about whether or not going to support Avalon and XAML is a good idea if your goal is nuke Microsoft and Redmond.
However
My goal for Mono, being an active supporter and a small contributor, is not to try and kill Microsoft. My goal is not like most slashdot zealots to wipe and replace Microsoft. My goal is to provide Linux with a platform for developers that they can and will enjoy.
The point is not to compete with the Java world nor to compete with the Microsoft world. The point of Mono is to create both a self hosting platform and a platform that will be somehow compatible with Microsoft.
The point of Mono is not the be 100% compatible! It has never tried to be 100% compatible. The main point of Mono is to create a self hosting platform.
People often argue that it would be better to implement our own kickass framework. Well, Mono is just that. Agreed they are filling in the specifications which Microsoft made. But Mono is doing much more than that. And the specification is not that bad at all. Why throw a way a nice specification just because you hate the creator of it? That doesn't make any sense. And I don't hate Microsoft, nor do most Mono developers (oh by the way, Miguel is not the only developer).
Hating Microsoft is foolish and stupid. You don't have to love them (hell I don't) and you don't have to agree with their marketing point NOR technical point of view (mostly for the marketing part I for sure don't), but that doesn't mean that you also have to ignore them completely. I even dare to say that you are a fool and an idiot if you do so.
I would very much support introducing support for Avalon/XAML in Mono if Avalon/XAML is a nice technology. And yes, it looks nice to me. So if it's possible to implement that technology (using Mono or using whatever) then I think that we as an OpenSource community should do that. We should, indeed, (re)implement it, at some point in time.
Not because we can then compete with Microsoft, thats not the point, but because we want to provide developers (and in the end, users of our softwares) with the best technology, the best platform and most choices.
Our users will have the benefit of not having to get locked in that Microsoft monopoly because WE recreated a part of that Microsoft-world.
Lets not forget that we are doing this because of the love of the art of programming, not because we HATE Microsoft. Thats what those stupid newbie Linux usies think why we do it. We love the art of programming. We love to show our art and the best way to do this is by making it public. And we, OpenSource developers, think that the best way to make things public is by licensing it using for example GPL, MIT or whatever OpenSource license.
Just like a lot musicians release their compositions for free, so that students can learn using their materials. I often compare such (classical) music with software code. The author thinks that it's art, the listener mainly enjoys it. But for a lot people it's art, okay?
For software developers, our code is our art. Our users don't give a shit about that code. They want to use our code. We want to distribute our art and show our skills. THATS the main reason why OpenSource exists. NOT because we HATE Microsoft.
Regretfully most people think we are doing this because we hate Microsoft. We don't. (And I speak for a lot OpenSource developers, I am confident about that).
There was a good reason to buy Windows XP - it is great. It's by far the best version of Windows since 95, and for people that were stuck with 98, or God forbid ME, there was a clear reason to upgrade.
Free iPods - now in the UK!
Microsoft is on top of that.
LUA is supposed to take care of that. And yes, it is a bit like Unix permissioning, but it does do some cool stuff, like provide each app its own copy of local files and even mock registry hives.
Mono means they can stay in their comfort zone, but still produce software that will work for people moving to Linux. You're not likely to change the minds of all those Windows programmers who are just doing a job because they are being paid to do it, but you can at least open a path of least resistance towards portability. Go Mono!
All those moments will be lost in time, like tears in rain.
Off topic, but ... Are there any examples of actual projects using mono that I could try out right now? (On Linux.)
.. Anything?
Web apps, desktop apps, utilities
I am slightly tired to this Longhorn debate.
Why try to re-invent everything that Longhorn already have?
If Longhorn is so good, then use it instead of wasting time re-inventing the technologies. And if a detail don't fit, then fix it on Longhorn.
I do. I use Windows XP hours on end at work. But I use GNOME on Debian at home. And I prefer GNOME over XP. Even though I'm on a 750MHz Duron.
And, in my opinion, it doesn't matter that I'm a power user in both OS's. I work as a student tutor at the local community college, and I see people completely new to computers coming into the lab every semester.
They don't find Windows intuitive. They don't find Office intuitive ("Where is cell B5?"). They don't find MS Paint intuitive.
The easiest thing for them to use is the Internet. And that's actually easier to use under Linux than Windows, since IE is absent under Linux. People get all these windows popping up over their screen, and they have a hard time doing anything about it.
There's a lot of people around who still don't know how to use a computer well. They go to community colleges to learn. Community colleges exist to serve the needs of bussinesses, and they have a tendency to swallow market speak. So market, damnit!
tasks(723) drafts(105) languages(484) examples(29106)
I see two possible options:
* Implement Avalon/XAML and ship it with Linux (with Mono).
* Come up with our own, competitive stack.
wxwidgets and python with a sandboxed execution stack using the already existing xmlrpclib.
Got Code?
Miguel makes good points but is wrong to attribute MS's dominant position to Linux apps being 'late to market', for example.
At the time of Windows 95, nobody could seriously say that MAC OS was not far better - stable, superior UI etc. So, even when MS had an obviously inferior product, they still won. And now, from XP onwards, they don't have have such a bad product, so what hope is there ?
They won because of their ruthless, illegal business practices. It's time to stop with the argument "we'll win because we're better".
"If we choose to go in our own direction, there are certain strengths in open source that we should employ to get to market quickly: requirements, design guidelines, key people who could contribute, compatibility requirements and deployment platforms."
Pity that he's obviously not been watching how most programmers actually do programming. Hint: most of them wouldn't know how to create a real requirements document if their lives depending on it. And read the requirements, and then develop real test cases that verify both functionality and coverage? Don't make me laugh.
Once upon a time there used to be two groups of people creating software: the analysts/engineers and the programmers/coders. The first group did the analysis, requirements, modeling and design; the second group converted it into code and punched it in. There was a reason for that, and those people produced some serious applications. Some of those apps are still in use today.
But, sadly, with the advent of the IDE it's now possible for anyone to be a bonafide Code Monkey, and just starting PAK'ing (programming at the keyboard) like crazy.
We're doomed, people. Submit to Bill now and just get it over with and save your passion for something more productive. Like sex.
Java failed because Sun assumed it was good to use on the web and it simply wasnt.
.NET (and Java too) come up with good, interoperable, solid ways to make this happen, web apps will be springing up in areas that you have never imagined.
No, Java "failed" [as a web app framework] because Sun never could put together an applets platform that was fast and produced professional-looking apps.
If you really believe there has never been any demand for fully functional applications running in a browser, your vision of the demand for apps has been far too narrow over the last 10 years. There was absolutely high demand for this type of application in 1995, and even more so now. Some isolated examples are coming closer and closer to this vision already, just making use of DHTML and proprietary browser enhancements. Good examples are the newer versions of Exchange Web Access and Hotmail, which are both coming closer to fully functional web apps with every new release. Once
More importantly, there is high demand for easily deployable applications in many business environments, and it's obvious that the easiest deployment is no deployment - something which is only accomplished via a universal tool that everyone already has - i.e. The Browser. Just because you personally don't see the need for a web app, does not mean that many thousands of companies with billions of dollars to spend don't have business needs for them.
People were sleeping at the wheel. In 1993-1994, Linux had the promise of becoming the best desktop system.
Miguel is fabricating some silly, alarmist, revisionist history with statements like these.
Linux was a lot of things in 1994, but one thing that is was not was a viable desktop. It was so lacking in the mindshare, number of developers, driver support and basic desktop technologies in 1994 as compared to today, that statements like this just make Miguel look like a silly idealogue.
This is the age old Microsoft mantra. Long before they've so much as written a line of usable code, before they've even tested the concept, they cry to world that their current vaporware will be the greatest thing to revolutionize the computing industry -- EVER!. There was NT, "The UNIX Killer" before whom the nations trembled, DCOM --- the ultimate framework that would redefine client/server computing, which more than anything made it possible to crash Windows remotely on other peoples machines. There was ActiveX --- Ooohhh --- another name for COM/DCOM/OLE/ATL which would change the web and make alive. Like DCOM, it was little more than a fancy DLL tied to the Windows registry. Lets not forget Windows DNA --- what disruptive technology was that? Then came "Next Generation Windows Services", which like DCOM was been morphed/recast by the marketing department into something more catchy. Yes, dear reader --- it truly is the lastest greatest world changing, paradigm shifting, not-to-be matched-or-conquered --- (trumpets blast) --- .NET. No really, we're serious this time, it's really going to change the world. This is going to be really good --- just wait.
So what's the reality. It's been three years since Almighty Bill declared to the world that Microsoft would make its software secure. Still waiting? Remember how the XBox was going to be the greatest gaming machine ever made --- a year before it was to be release? Well, I see playstation still has 60% of the market. I also hear that XBoxes have been know to catch the carpet on fire.
Maybe I'm too old. At the ripe old age of 33, I've smelled enough MS BS for a lifetime. The only thing I do now with this kind of news is use it to compost my wife's azaleas. I've yet to witness The Unix Killer, trustworthy computing, DCOM in my life, and somehow I doubt Longhorn will change this. I am quite happy with that "Cancer" called Linux and GPL software, that just three years ago was never going to take off. Yep, I'm shaking in my boots.
This is only half true. As the 'computer person' in the family I get all of the 'my computer...' calls. Most of these calls are related to virus activity and or crappy performance. The most common being performance issues. 9 times out of 10 this is related to LOTS of spyware. Every machine I sit at after removing all of the spyware and setting up adaware to run every so often I remove all links to internet explorer from the users desktop and start menu and replace them with Mozilla. I name the link 'Internet Explorer' and replace the icon on some machines because people do not know what to do in any other case. Once this is done I always get comments like 'The Internet is so much faster now!' and once I tell them about the new 'feature' of tabbed browsing it is a done deal. 100% of these users are STILL using Mozilla. As I am sure all of the readers here know its all about what is PRESENTED to the user. As long as MS continues to bundle IE it will be the dominate browser. If we do not want that to be the case then start removing it from user land and putting mozilla in its place.
of course failed to find an agent
Why "of course"? Some kind of conspiracy theory?
So maybe the Mozilla team should consider creating a XUL plug-in for IE then. Is that feasible, or are there technical quirks preventing that from happening?
- Implement Avalon/XAML and ship it with Linux (with Mono).
- Come up with our own, competitive stack.
There is, of course, a third option. Move all your development time and efforts to supporting and using Java on Linux. Java is the technology Microsoft is trying to emulate and squash withCan you recommend a couple of sites to learn about Linux text-to-speech and/or voice recognition, especially X integration?
You can already "embed" XUL in IE of course, by having the user download the Gecko ActiveX control and effectively embed a renderer within a renderer, but that's a cheap hack and has severe performance implications.
To be frank, I'm 100% not convinced that Avalon is going to be as world changing as Miggy predicts. I think it's especially rash to be starting internal projects even if they are "thought only" to develop a competitor.
Miguel thinks Avalon will be great, because it will let you deploy applications via the web browser that use native widgets, and be nicely integrated with .NET and so on.
But ... but ... but ... Microsoft did this years ago (minus .net). Or am I really the only one who remembers the version of Outlook implemented entirely using DHTML/HTA (which produces native widgets). I can't remember the codename, but the project was scrapped. The benefits of running Outlook inside IE just were not compelling enough to overcome the performance and other problems.
I'm not denying it'd be useful. The long term UI goals for my own packaging/installer project are that the user should be able to launch (and implicitly install) software simply by clicking on an icon embedded in a web page. As far as the user is concerned then the effects would be the same, so the real differences lie in how the developer sees things.
In the Avalon world view, the developer creates widgets on a canvas (AFAIK), ties them together with .NET code, and then .NET CAS allows you to launch it from a web browser without fear of it doing nasty things to your machine (which is a massive oversimplification, but oh well).
In fact, we can do this sort of thing today, with technologies that are either here or just around the corner. SELinux allows you to effectively sandbox native code to a fine degree, similar to .NET CAS except enforced at the kernel level and not by a VM. It's not just a set of kernel security checks - it's actually a security architecture with an exposed set of APIs which allow people to use MAC security features to a high level.
I don't know enough about .NET security to know how it compares, but SELinux policy is easily distributable in the form of text files and allows you use native code, which runs directly on the CPU without the overhead of a VM and huge set of managed APIs.
So, if you can download some native code and correctly sandbox it, you have the start of web app deployment. XAML appears to bear a superficial representation to Glade (note: NOT xul) but using a more customised and therefore human friendly schema.
I say Glade and not XUL because a Glade file is, in actuality, not an UI description at all. It's really a persisted GObject tree that libglade uses along with the GObject reflection APIs to reconstruct the GUI at runtime. I have read that XAML despite appearances is similar: it is a persisted .NET object graph.
So, I think if Miguel starts from "what user experiences does this technology allow" and work backwards, he'll find we already have the basics in production. Sure, they need to be improved and tied together, but they are there nonetheless.
Finally I think it's wrong to say that the reason desktop Linux didn't happen in 1994 was because people were "sleeping at the wheel". The fact of the matter was in 1994 Microsoft already had several thousand people working on Windows full time, whereas desktop Linux had .... none.
Really, I think a simpler explanation is just that MS had a monopoly pumping cash into their development teams, and Linux did not. Its falling behind was therefore completely inevitable until it gained enough momentum to move as fast as Microsoft do.
Here's a question. How do we copy Microsoft and get our "working something" out the door first? Do we go back in time?
Here's another question which nobody on the "We must clone Microsoft's products at all costs" lobby has ever satisfactorily answered: how are we contributing anything to the world if our product is just a (poor, it has to be poor, because Microsoft's technologies are not lumbered with having to run on a platform that was never designed to run them) clone of something that already exists? Do we improve music by producing "free" versions of Brittney Spears and the Spice Girls? Do we contribute something new and wonderful by making a movie with the exact same plot as Terminator 3, but with even poorer acting? Do we ensure that everyone has something that caters for them by spending a lot of effort cloning the writing style of Jeffrey Archer and writing predictable thrillers, then redistributing them for free?
As long as Microsoft defines the product, Microsoft will be ahead. Icaza ignores this because Icaza likes Microsoft's technologies, they suit him, he lacks the imagination, will, and talent to produce anything better than what Microsoft produces, so he's content to spend the rest of his life in perpetual catch-up, pretending that he will in some way be able to produce something better than his mentor and rival without ever knowing ahead of time what it is that mentor and rival will be producing and so being unable to produce his clone before Microsoft's original.
The truth to the statement "The central point was that paying too much attention to Microsoft simply allows Microsoft to define the game. And when Microsoft gets to define the game, they ALWAYS win." is self-evident. You cannot both follow and lead.
You are not alone. This is not normal. None of this is normal.
MS will win because it looks for problems it can solve for customers with its technology. MS employees are not looking at Linux and going oh look at that feature we need to counter it. Or no problem we already have done that.
Instead they look at the market and say, how can we solve someone's problem. A great example is thin media clients. Linux could have dominated this market. Linux is a robust OS that just runs. It has a low to no cost for deploying to millions of homes. The HD1000 from ROKU (http://www.rokulabs.com) is a great example of what is possible for Linux in this $100+ billion industry.
However, Linux is squandering away the opportunity. MS came in to the marketplace and said to the hardware manufacturers here is a complete solution just install. To a company that cares more about selling hardware than software the choice is clear. Pay MS and design the hardware to run MS technology (especially when you have multiple hardware vendors saying here is the base platform already designed for you). The consumer electronic companies make money by selling hardware not software. Anyone who says to them here is a complete and working system just build the hardware will get there attention.
That is why MS wins. They solve problems; they don't just invent technology for technologies sake.
Rabi Satter
The main reason Java failed on the desktop was that there didn't used to be a Java Web Start program. But now we have one and it works great.
Also, it used to be difficult to install the Java VM, and the Microsoft version that came with Windows was a buggy piece of crap. However, that's been solved too. Visit www.java.com in IE and click download. It's as easy as installing FlashPlayer.
Other issues: AWT was too sluggish. Well, the new Swing UI is pretty slick. There was no good Java IDE. Now, Eclipse kicks ass and is the best IDE evar IMHO.
And as for Applets being mostly useless, that's not true. If you sign your code you can do anything with an Applet that you can do with a Web Start application.
We have great success with Java on the desktop. The only snag is that we have to tell our customers to download the Java VM, but once they've done it, everything's good.
Time is an illusion. Lunchtime doubly so. --Ford Prefect
There are plenty of others. For Speech synthesis, you are probably looking at Festival. For Voice recognition, you are probably best off looking at IBM Viavoice for linux. GNOME has gone a very long way with the Accessibility toolkit and will continue to push down the accessibility path - for example, take a look at Dasher for an interesting app to aid writing for impaired users. There is a lot more on GNOME Accessibility to read.
Cheers,
Toby Haynes
Anything I post is strictly my own thoughts and doesn't necessarily have anything to do with the opinions of IBM.
First of all, you have to decide who you're talking about. There are server based code(ie. web development, aka think client) and client side code (ie. fat client). I develop server based code. We use Java. We use Java because our customers use so many different platforms. We've deployed on solaris, mvs, linux, windows, and many more. There is not even a remote possibility that we will be switching .NET for these types of applications in the near future. The only people who would switch their server code to .NET would be people who are currently in the VB.NET/ASP.NET world already.
.NET? Your photo editing software? I don't know, I can only speculate. A direct comparison between OSS and commercial/microsoft versions of a product reveal that in most cases the OSS version is more secure and has better features. So why, oh why, do people not use the OSS version? Simple, marketing!
The client side (desktop) is the area where all Miguels comments seem to be directed. Will your word processor of choice be written in
You see, software developers work on projects. And projects ARE NOT PRODUCTS! You can have a successful project, which may not be a successful product. And as microsoft shows, you can have an unsuccessful project, which is a successful product. Projects become successful products because of good marketing. OSS has little or no marketing, and this is the fatal flaw. If only apple could help market some OSS projects we could see just how successful they could become. Think about it, if you saw an ad for the "Sexy, New, Feature Rich, Gimp project"(note that a name change would be mandatory for this project to be marketed, project vs product). Now put this ad in Cosmopolitan magazine (this is where you see ipod ads...). Put it everywhere. Make it sexy, make consumers, that's who we're really talking about here, want it.
Many of the developers on these projects are not going to like this. Nobody wants to "sell out" their project. But if you're after the client side market(aka desktop), then you're targeting consumers, not developers.
mp3's are only for those with bad memories
The truth to the statement "The central point was that paying too much attention to Microsoft simply allows Microsoft to define the game. And when Microsoft gets to define the game, they ALWAYS win." is self-evident. You cannot both follow and lead.
That is specious reasoning; Microsoft gets to define the game regardless. No matter how much we innovate, the pain of migrating to another platform keeps companies on Windows. If we created the next killer app, Microsoft would have plenty of time to copy it before people started to migrate en masse.
The only way to ease the pain of migration is to make things work. Most companies' infrastructure is far too thick to be able to migrate to a whole new platform in one giant leap. So addressing Windows compatibility is critical before many people can even consider Linux.
That said, I agree largely that a single project can't lead and follow, but GNU/Linux is not one project. If you are arguing that resources spent copying Microsoft are wasted, then I think it is only your own time that is being wasted, since open source developers work on what they want and will never all agree to one ideology.
When originally I heard about Mono I was skeptical. Then I met up with Miguel had a talk to him and was optimistic. There were some posts of his that made me upbeat about Mono. Now ever since Novell bought Ximian I am really skeptical again.
.NET clone. Mono will never succeed and it will fail miserably. Nobody can compete or be compatible with Microsoft, just ask Mainsoft, Bristol, and other companies that licensed Microsoft technologies.
Mono SHOULD NOT be a Microsoft
I am amazed that people think it is in Microsoft's interest to build cross-platform application. Microsoft has said, time and time again that it is not in their interest. Microsoft has their own operating system and that is their interest. So what I wonder is why people keep thinking it would be good to run Windows Apps on Linux.
Wine, and CrossOffice are hacks until more applications are ported. When I use my OSX box, or my Linux box or even my Windows box I look for native applications, not emulation. Native apps run faster, better and are more stable.
Mono should go back and focus on doing their own thing again. Just like the Jakarta team focused on building good Open Source applications.
"You can't make a race horse of a pig"
"No," said Samuel, "but you can make very fast pig"
Miguel's crusade to badly copy where Microsoft has gone before isn't really that productive and it has produced rather a lot of sloppy, unfinished, unpolished software that has more promise than usefulness.
I desperately want this not to be so, but it is.
Microsoft have an important ally in Miguel. It is not necessary to announce vaporware for Linux to frighten off the competition since everyone is already waiting for applications like Evolution to stop sucking so badly.
Am not fabricating anything.
In 1994, the desktop was not a GUI desktop, the
desktop was mostly a command-line universe both
on DOS-based systems and Linux systems.
Linux did have an advantage: multiple virtual
consoles, real multi-tasking, tcp/ip stack
bundled, nfs, file serving capabilities, and
DOSemu with compatibility with the past.
I have to say, way better than DOS + pile of
device drivers and Windows was only starting to
be used with very few applications. Windows 3.11
was out, with really few applications.
Miguel.
I liked the sentiment of your post, but I have two minor criticisms:
/. for years, wanting to contribute (but were unwilling to go back to C) will help build the next Linux desktop.
1) Mono is not the new Wine. The mono implementation of WinForms may be (at one time they talked about binding it to wine). I see the main use of Mono to be building Linux applications that only *happen* to work on windows. At first they will use GTK# and later some new better UI toolkit (Avalon# ???, XUL# ???)
2) Mono opens the doors for many more programmers to contribute to Linux, but not just unwittingly. When (if) mono become commonplace on the Linux desktop, Window programmers and Java programmers will flock to the Linux desktop. Thousands of people who have been reading
I for one welcome the change. God knows there's plenty of work.
Windows 2000 is not a good OS for the home. My memory is sketchy since I wans't using it much (except at work) but let me see...
What version of DirectX did Windows 2000 ship with? I think it was way behind or something. Games and other multimedia apps weren't very good...
What was the cost? If I remember correctly, didn't Win2000 cost more than Win ME and Win XP?
Win2000 boots up slower than Win XP.
Win XP has better sleep mode, and other power consuming features.
And so on...
Overall, Win XP is much better than Windows 2000 for the home user... Don't know about servers though... In any case Win XP is very similar to Win 2000.
Sivaram Velauthapillai
Sivaram Velauthapillai
Seeking the meaning of life... @slashdot of all places
A: Security (you know Microsoft code will be riddled with holes here).
I can write insecure software on Linux just as fast as I can on Windows.
B: Realiability
See above. I've had Windows boxes that are very stable (a year of uptime).
C: Peer review, I, John Q random engineer can verify it.
When was the last time you looked over every line of any OSS package?
D: Speed, basically Windows is bloated and slow.
Funny... my benchmarks in the past don't show this. Compilers from Microsoft (which are what most folks use) tend to do much better optimization that gcc (which is what everybody uses on Linux even though Intel compilers are much better and also free). Some benchmarks I have run on FP intensive code have shown to be 2X as fast on Windows with Microsoft compilers than with using gcc.
Also, X tends to be slow as a pig even on my high end graphics cards.
E: Continuity, a user of the original Unix would be able to navigate and use Linux desktop in hours, you cannot say the same thing about Microsoft software.
And exactly how many of those folks are around? You may not say the same for a Unix person migrating to Windows, you mean? Also, because XWindows folks tend to customize their desktops a lot, I would say that X users attempting to use one anothers desktops is a large hurdle.
Eventually people will get tired of continuously shelling out for the same regurgiatated Windows 95 core functionality.
Obviously you haven't used Windows since the Windows95 era, which was 7+ years ago, which is why you posted all of the above outdated noise.
It will be the likes of the Chinese, Japanese, Germans and developing nations, which will, break away from Windows with force, it's already happening...
Which oddly enough has much less to do with any "technical" reasons other than political and economic.
1. Other countries don't like seeing their money go to the USA. (economic and political)
2. Other countries are possibly afraid of "back doors" in an OS that is provided by another country. (political and defense)
3. Other countries would rather put money into projects in their own country or region, such as a Linux distribution (political and economic).
4. Other countries want to do whatever they can to switch control of things from a central controller to something that they have more control over (Microsoft controls a lot, knock them down a notch or two and make us stronger) (political).
This is very evident in such things as AirBus, who is 1/3 subsidized by the European community in order to compete with Boeing and other US based aircraft manufacturers. This is done to bring the economy of those industries back to Europe and politically, to break the dominance of the US aircraft manufacturers. The same is happening with Linux distributors because Linux is an OS that is "Open", already underway and working, and is easy for any country/region to support a "local" enterprise to get started.
Provide 'analagous' and seamless cloned functionality for less and Microsoft has *no* market.
Yeah... still waiting for anyone (Microsoft, Linux, or otherwise) to provide that.
Disclaimer: I am a Linux developer. I have used Unix/Linux since ~1986 in a variety of flavors. I have yet to be "satisfied" by any Unix/Linux or Windows offering. They all suck, it's just a matter of picking the least sucking one for what you are trying to do at any given time for a given problem. At times in the past, that has been any number of Unix/Linux flavors, Windows, or other embedded platforms.
Why would a bank use XUL for an application (forcing all thier users to download mozilla), when they can let them run native Longhorn appps from thier browser without any installation?
Mozilla has the GRE (Gecko Runtime Engine) which is all that is needed to execute XUL apps. The GRE can be loaded without the browser. Last time I check I think it was a little under 10 megs which is not too bad since 1 GRE can support multiple XUL apps.
Linux won't overtake Windows for a while (I'm talking about home market only--server is another story). Windows just provides too many features that Linux lacks. Windows is also much easier to use. For instance, how easy is it to change video card drivers (or for that matter any driver) in Linux? It is very difficult for causal users. I think 90% of the home users won't get past, say, Nvidia's 3rd instruction to modify the XFree-86 configuration file. Or how about when something goes wrong? Linux is not as foolproof. For instance, if there are problems with the filesystems, the error/fix messages generated during boot-up would give a heart attack to many users (the messages basically say something like 'you could lose all data if you proceed' :) ).
Lastly, Linux does not have enough software. This will seriously prevent adoption for a while. You can find basic software in Linux but it's always those one or two special ones that make it difficult. For instance, if you are using a particular tax software, you may not find it in linux. Or you may be able to find replacements for everything but MS Encarta (encyclopedia). Or, you find everything in linux except that digital camera software that you cherish.
Sivaram Velauthapillai
Sivaram Velauthapillai
Seeking the meaning of life... @slashdot of all places
A Microsoft clone of GNU/Linux is as bad a thing as a F/OSS clone of Windows. Clones are bad. Choices are good. Clones remove choice.
Remember the late eighties? You had Amigas, Macs, Atari STs, PC Clones with GEM, PC Clones with Windows, and those were just the "mainstream" platforms. You had choices. You could chose a computer that actually suited you. The different manufacturers did things in different ways to suit their audiences. There was more than a nod to the Mac in all of the above, but not so much you could safely argue most were clones of it. That was a good time to be in computing.
The F/OSS communities can be leaders in the industry. It doesn't have to replace one monopoly with another. It certainly doesn't have to submit to a monopoly mindset, as its leaders do today.
This is a justification for cloning Microsoft, not something that addresses the issue of Microsoft's ability to define the game. And it doesn't answer the fundamental "what's the point of creating something that isn't a choice" issue I raised in my original. I'm arguing that copying Microsoft is fundamentally damaging. The cloners are more interested in the idea that something called "GNU/Linux" will become popular and that "Microsoft" will not than producing something positive. They don't care what gets called GNU/Linux, as long as it "takes over the world". If it's a lousy, security hole ridden, irritating, poor clone of an operating system that was never any good to begin with, that's fine, as long as the name wins out.This shouldn't be a war against Microsoft. This should be a war against a lack of choices, and against proprietary software. Both are inherently undermined by the cloners.
You are not alone. This is not normal. None of this is normal.
I agree with Miguel here,in '94/early '95, at least in the UK, there was talk that if Windows95 slipped much more, OS/2 Warp might take over! That's how weak windows was (then known as Windows for workgroups 3.11 I think).
Follow Miguel, follow Microsoft... there's not any difference except in the end, one may have more of a surprised look on their face than the other. I can hear, "Oh... well.. I never saw that one coming." But in reality, I think Miguel is smart enough to FULLY comprehend what will happen... and that's what is really scary.
Miguel would say that we're all asleep... are we?? I wonder who really has their eyes closed on this one.
Check out Avantbrowser. It's a replacement "front end" for IE, supports tabbed browsing, popup blocker, ad blocker, script blocker, flash blocker, etc, etc, etc. Ctrl-N (or middle-mouse click, or mouse gesture, or however you want to open a new tab) works as you'd like it to (and me, too). As for wrong URLs hanging for 10-20 seconds, that's an oddity. I usually just hit Esc to stop loading the page.
Which is more painful? Going to work or gouging your eye out with a spoon? Find out!
http://www.workorspoon.com
I think your memory is a little different from mine. Certainly, DOS was the PC king. Almost all games came out for DOS. Lots of various bits and pieces and utilities came out for DOS. DOS was where a PC hacker got work done. DOS was where we computer-type boys and teens mucked around on.
But for the other 90% of computer users it was a different story. Windows 3.1x may not have had alot of software from the hacker's perspective, but it was a smash hit in the home-user market despite its limitations. It is Windows 3.1 that sent Amiga to its grave, battered the Mac platform, and for the first time had masses of ordinary people seriously considering buying a computer. It was cute, and although it was not very intuitive, it was simple enough to learn for the more intelligent 1/4th of the population who considered buying a PC back then. Microsoft's marketing at the time was the best it has ever been. They perfectly understood the paradyme; their "Microsoft Home" subbrand (that finally got mothers wanting computers for their children) might've sold over 10 million PCs by itself. Which is about how many people use Linux today incidently.
In 1994 the desktop was a GUI desktop -- for everyone except hackers. The GUI desktop had easily arrived circa 1992 for oridinary users. Hackers, on the other hand, didn't seriously get into the GUI thing until KDE arrived. Before then, the Unix GUI was just a glorified text and widget terminal.
Linux did do well for what it did. It was a geeks' perfect toy. Its technical features were excellent. It did expand to its full potential for what it was capable as -- as a hacker's _text-based_ Unix system.
As much as Linux had an advantage over DOS and Windows from a pure technical point of view back then, it sucked as GUI desktop system. By and large, it still does, as much as Linux users protesth. That's why hardly anyone uses it. This is the cold hard reality.
The only way Linux could have succeded on the ordinary user desktop would have been if it didn't use X-Windows, didn't have a heap of semi-broken widget sets, etc etc etc. I don't need to explain this. We all know what the problems are. If Linus had in 1994 decided to personally oversee and start from scratch a new, unified, coherent, legacy-free 'official' GUI project for Linux, then Linux would be used by over 100 million people now.
After seeing Nat speak at the Real World Linux Conference a few weeks ago in Toronto, I decided to check out Dashboard. From what I can tell, no it's not usable. It's clearly CVS-quality code, but that's not the reason it's so difficult. The barrier to entry for compiling Mono apps seems VERY high: you'll need to compile gtk-sharp, gconf-sharp, glib-sharp, gnome-sharp, evolution-sharp, as well as all their esoteric dependancies.
Presumably these dependancies will come out of CVS and will be packaged with GNOME, but for now "mono-izing" your computer is a pain in the ass. I would only suggest installing Dashboard if you have lots of time and patience. Otherwise, wait six months or so.
What kind of computer are you using? Saying something is slow without specifying the hardware you are running it on, is like saying I have 500 units of cash in my pocket: what lire? euros? cents? dollars? Yes. JEdit is slower than vi. But I am using at least 5 java apps continuously. I really like the fact that I KNOW I can switch computer and still use them. So the slowdown is not a big problem given that functionality. And since we are speaking of mono, are you suggesting that will be faster?
Well, C coding *IS* dead, or should be. Whether OSS is the future or not is debatable, but non-object oriented, non-exception handling, non-bounds checking languages with hand-rolled memory management are on the way out. They're inefficient to program in and nowadays have little to offer in terms of performance. And thanks to the unsafe block, there are ways to bypass even the smallest performance hit by removing all these safeties.
.NET initiative in a cross platform manner. There's a subtle difference, but it's an important one: de Icaza's methodology takes the newest strategy from the for-better-or-worse market leader and makes it ubiquitous, instead of trying to make a name for himself with a brand new strayegy. From a risk assessment point of view, there's a much better chance that .NET will succeed than any of the dozens of competing intiatives in the OSS community. And there's less work involved. We the power users may not want to get both feet in bed with Microsoft, but for a lot of companies out there it has proven to be a very valuable strategy. Miguel's trying to give them a means of keeping one foot on the floor, to tap the ubiquity of windows while maintaining (or in many cases, gaining) cross platform compatibility.
.NET. MS couldn't say that openly, of course.
As for XUL...i can't see why anybody who touted the life of C could also praise XUL. XML is a nice idea for encapsulating data in a hierarchical, human readable format, but it's a bad bad BAD idea for user interfaces and anything else where you want INSTANT access to data. Parsing -- or should I say compiling -- all those words into language a machine can understand wastes time. Sure, it makes sense for a handful of widgets (like a web page), but what if you have an application that loads 300-500 per form like most of the apps I deal in? Not only do you have the rendering overhead, you also have the XML parsing overhead for each of them. I'll stick with JVM and the Windows Forms frameworks.
As for "catching up" with Microsoft...de Icaza's point is that while Linux is treading water with its own kind of uniformity and platform cross compatibility, trying to make inroads into Windows apps, de Icaza's aiming to replicate the
Which I'm sure was the whole thrust behind standardizing
Hey freaks: now you're ju
I do not think you read my whole message,
because I stated that there were two options:
to implement Avalon, or to build our own.
We are in the process of specing out what
ours should be (the platform we call
"salvador").
Miguel.
But many of you have ignored the reason why .NET is going to be so popular - it is going to be so damn easy to develop applications. Literally, everything will be taken care of for you. Want to link to a database? No problem, in 20 minutes you will have a database application. Most people here will probably not want to use it, but we aren't the ones making most of the decisions. If a company sees an opportunity to hit 90% of the market (which it wants) and develop the program in half the time, then it will. There is no arguing the economics of this. .NET is a formidible beast and yes while most people here say it sucks and they hate it you don't seem to realize most people don't care. Or if they do care, they are willing to set aside their caring long enough to write a program in .NET, which will take all of a few days for a lot of apps.
http://www.rustyrazorblade.com
Sure enough, Microsoft has DEDICATED people reading this stuff. Access to it is just a click away. Market strategy is all about surprise. So I'm proposing a new movement. Open Source, Closed Strategy (OSCS). Seriously.
"In order to increase developer productivity, Avalon will rationalize and reduce the number of APIs in the Win32 stack from over 70,000 down to 8,000."
No it won't. All the existing APIs will still be there, because existing applications use them, and if Microsoft was interested in breaking existing apps for a good reason they'd have done it already.
Avalon will add a new 8,000-element API to the 70,000 already there.
[more blurbs]
Sounds like Cocoa.
Get it overwith. You are obsessed with Microsoft. Every time you give a speech its about Microsoft, every product you seem to work on is a clone of something made by Microsoft. You are more worried about what Microsoft is doing than you are about what Linux is doing.
Hey, welcome to Slashdot and its community...
Sure it would be great to get many applications developed on .Net to run directly on Linux and BSD but I find the contrary to be true. Getting Linux apps to run on .Net means they will have a shot at running on Windows.
Its theoretically a two way street. Evolution on Windows? Pan on Windows? Sure leveraging all of those Windows applications on a platform of your choice is an interesting thing but everyone seems to neglect leveraging Linux and BSD applications onto Windows!
This will be interesting. I'm not going to be against Miguel or Mono but I'm not going to be against Cringley either. Cringley's point is that you can't lead if you are always following. Miguel's point is that its stupid to ignore such a good piece of technology. Why do people assume one is wrong?
...and found it to be less than enlightening. The first part reads as a self-serving defense against Cringely's (rather obvious, I thought) observation that *you can't play catch-up against Microsoft, because you will lose*. This painfully evident observation applies to everyone; if MS makes the rules, MS will always be at the forefront of whatever it is that it gets to define, and no one playing tag-along will ever be able to catch up with them. Miguel and his efforts are no exception to this, despite what he seems to think. He makes the mistake of assuming that he's different - just like all the other companies which thought the same thing, and were driven out of their markets (or nearly so) by Microsoft.
Cringely was right. Miguel is wrong.
The second part is based on a faulty assumption, i.e., that most Linux users care if about taking the battle to Microsoft and 'getting Linux on the desktop'. Fact is, most Linux users could give a shit one way or another, and aren't interested in seeing their OS used as a vehicle to launch a crusade against the evil empire. Never have been, never will be; this 'crusade' mentality belongs to a tiny, but very obnoxious and very loud, minority. One which I heartily wish would shut the hell up, move on to the next Big Thing(TM), and leave those of us who actually code for and use the OS for our own satisfaction the hell alone.
It's just as Cringely said. The best thing to do for Linux is to simply ignore MS altogether and continue coding what we want to code, when we want to code at, in the way we want to code it. If more than that tiny minority of us actually begin to take this crusade bullshit seriously, all we'll end up with is a second-rate Windows clone that whores itself out to whatever blithering idiots scream the loudest and whine the longest.
And in any event, what do these crusaders think they're going to accomplish, anyway? Even if they manage to drive MS out of business (not in this lifetime, pal) Bill Gates will still be one of the richest men in the world - richer than any of the crusaders, and laughing all the way to the bank. So will his cronies, and so will the smart investors. The only people who'll 'lose' the war are a bunch of average-Joe schmucks who work for Microsoft, or who invested in Microsoft and didn't manage to pull out until it was too late.
The folks who created the 'evil empire' have already made their money, and nothing the crusaders do to the company will change that fact. Those folks will always be richer than the crusaders, and will always be laughing at their trite vitriolic dialogue - no doubt while sitting on a beach in Tahiti surrounded by beautiful (if purchased) women. To them the crusaders are nothing more than sad little fools worthy of little more than contempt.
And they're right.
Max
My god carries a hammer. Your god died nailed to a tree. Any questions?
MIguel, in his usual way, is about halfway right, I think.
.Net will Take Over The World(TM). From his perspective as a .Net implementor on Linux, he obviously sees it as the best thing since Corona beer and tacos. Those technologies will surely become very popular in the Windows world, and I'm sure that a good deal of companies that are currently within the Windows loop will make heavy use of local Web applications a la XAML.
.Net and XAML on Windows machines since the idea that Mozilla will get it together in a reasonable amount of time to get their engine to render anything in the way of the Avalon engine is probably expecting too much.
He is definitely right that MS won a lot of its marketshare by simply bundling stuff with the OS and by having enough money and time to survive mistakes that killed competitors (XBox, WinCE, Plug and Pray, Bob, J++ etc).
He is only halfway right that Longhorn and XAML, Avalon and
But, as has been the case before, it's only half the picture. The other half of the picture is that those people who see it as critical to have their web applications be compatible with the myriad different Windows OS versions, the myriad different OS types right across the board will still use Java/PHP etc for server based apps and keep the frontend in the browser. The XAML local web applications are very similar to Java Webstart in concept, but will find it only marginally more acceptable in the real world, for purely compatibility reasons.
Granted Java has been an unmitigated disaster client side, with Sun having screwed up by introducing the white elephant known as swing and thereby permanently giving client side Java the reputation of being slow, even though this is no longer true with modern CPUs. This hole will probably be filled by
And the price/performance and price/freedom of implementation benefits of Linux are truly starting to find adherents across the world in a serious manner.
In the end it will probably be that Windows will provide the better experience but that Linux will provide the lower cost and "be good enough" very much like Windows 95 was compared to its competitors.
I beg to differ. 3D icons are a great idea. They can be scaled without pixellation and will look the same regardless of size. Furthermore, development of 3d acceleration has vastly outpaced non-overlayed 2d acceleration...meaning that displaying, scaling and manipulating simple 3d widgets might take less processor time and resources, as it can be pushed off onto the GPU. Repaints and updates could be performed with little or no need to interrupt other logic, meaning a snappier interface that won't have to block while waiting for single threaded operations to complete. You'd also get "free" anti aliasing, free animations, free shading, free state change on click, etc...
4d icons are also a nice idea, because they can display more information than a stationary icon. Think of the bouncing dock icons in Mac OSX or some of the many OSX icons that change their state periodically to inform you of other information. iCal changes its icon to whatever today's date is. AIM changes its icon depending on whether you have a message waiting or not. Windows minimized to the dock display the contents of their window. Even The Gimp makes use of this, changing its icon to a small version of the window's contents (does this under Windows too).
You're right...an icon IS a picture that represents an application or idea. But syntagmology is hardly "simple," as Ferdinand de Saussure would tell you, there's a LOT more information inherant to a sign than just its name. As long as that Save icon does more than just accept a click, there'll be call to make the icon do more than just sit there. It should react according to what you told it to do.
Hey freaks: now you're ju
The comment I responded to implied that we shouldn't lead, basically by wording the idea negatively (if we create something original, MS will copy it - well, duh! That's what leading the industry is. You should worry if nobody's copying what you're doing, not if people are.)
This is just semantics. I agree that Microsoft products are better because of the many innovators that fly under the mainstream radar. My negative wording was because Microsoft will implement a feature only as well as it needs to maintain it's monopoly, which generally means only to a marketable level, not a technically robust level.
The problem with your suggest is that Python has no sandboxed execution stack (bastion/rexec has been removed as of the 2.2 branch because it was fundementally insecure.) There is a lot of discussion about what to put into Python to replace this feature set. Personally, I favor a capabilities approach but Guide seems to disagree so we will see what happens.
Either way, only one of your two tools meets the required specs. Try again...
Longhorn always been targetted for late 2005/early 2006. People who still refer to Longhorn as "vaporware"--even with the PDC build and endless technology demos--are buying into a mindless hatred for all things Microsoft and ignoring the real existence of the technology that will be coming out and permeating everywhere.
.NET will change things. It's why they're completely replacing Win32 with it. I have a feeling you haven't really examined the Longhorn tech all that much and have only read some marketing hype that you subsequently dismissed. Surprise, surprise, companies market their products as the greatest things ever. But Longhorn is actually a real overhaul of Windows, from the display technology to the filesystem technology to the runtime libraries and more.
Sorry, Longhorn is not vaporware, and there has never been a release date delay because there never was a release date. And, yes,
Miguel is smart in recognizing the inevitability of Longhorn...Slashdotters want to dismiss Longhorn because deep down they know it will arrive and take over.
Miguel, what do you envision as the channel for getting Windows users to adopt Avalon? A default install is pretty hard to compete with.
Microsoft followed a development path with DOS/Windows which followed Moore's law. That is, they released software which had hardware requirements meeting the machines of the day. Thus Win3.1 could run reasonably in 4-8 megs, Win95 in 8-16 megs, WinNT in 16-32 megs... etc.
But in the early days, Linux had hardware requirements which far exceeded the capabilities of the common desktop. This was part of the great debate, and as an experimental platform it made sense to go this route. But it wouldn't have helped it to succeed as a mainstream desktop.
It wasn't until really probably the era of the Pentium II/III when desktops started coming routinely with 64 Megs of RAM that things had caught up with Linux resource needs. So I don't agree that Linux had a chance on the desktop in 1994, no moreso than OS/2 had, probably even less the case as OS/2 had numerous large scale deployments in Fortune 500 companies. Even then, factors contributed to prevent it from taking off.
Linux did have a chance right around the 1999 time frame to make signifigant strides, this mainly due to a weak spot in the market left by the delayed delivery of Windows 2000 to upgrade the slagging NT4. But since the release of Win2k, there has been no compelling technical reason to deploy Linux in either the desktop or server realm. I think Miguel is correct in that this situation is going to become even starker with the release of Longhorn. There will be a substantial gap between the capabilities of Linux and Windows.
I do have to applaud Miguel for his technical understanding of the issues, and his work on Mono and other technologies. It would be great if some day it was as enjoyable to develop on Linux as it is on Windows.
I have trouble understanding your contradictory stances. The difference in performance between C and, say, Java is unimportant, but the time to parse XML is a deal breaker? I would rather parse 10 pages of xml on startup than wait for the JVM to fire up. Of course, if you are arguing for C++ or some other compiled OO language, there is still more of a performance hit compared to a lower level language than there is in parsing a few extra lines of text. (Just to clarify my own position, I favor low level non-OO languages and despise XML...)
Boys from the City. Not yet caught by the Whirlwind of Progress. Feed soda pop to the thirsty pigs.
First: looking at things from developers' perspective does not make any sense. These days the majority of PC users are not developers.
Second: MS won everything mostly due to their absolute commitment to backwards compatibility. It allowed them to survivie during Sux'95 days of BSOD and will allow them to survive the present security fever.
But that's nothing. What's important is DRM. Here is a probable scenario for the next decade: 1. DRMed Longhorn PCs become commonplace. 2. Most of the commercial content is accessible via DRMed PCs only. 3. Linux' chances to win a consumer desktop are dead once and for all.
I mean, commercial websites will not be accesible from nonDRMed Linux/Gnome/Mozilla PCs because those web sites will be happy to use DRM for: 1. blocking ad blockers 2. blocking "helpful" slashdotters from violating their copyrights.
And chances for a GPLed kernel to become compatible with that patented MS DRM stuff are zero.
Right?
And --POOF-- Slashdot disappears in a cloud of irony.
-
Inventor of the term 'pardon my French'.
I would rather parse 10 pages of xml on startup than wait for the JVM to fire up
Interesting point. Obviously, if you're starting up a lot of programs all the time like most UNIX coders, you don't want to open up a new JVM instance for each. I don't deal with piping or porting, so I didn't think about it...most of the programs I write are opened at 8:30 am and not closed until 5 pm. What you consider a program is more like a function to me...hence why you don't necessarily need OO.
In fact, I think that's the key to most of the pro-C, anti-OO sentiment. You're thinking of programs as things that start, perform a set process, and then end. I think of programs as things that start, and then do whatever the user tells them to do. To you, three separate user functions are performed by three programs called by a flow control program. To me, they're three separate functions of an object, or three functions in three separate objects, or one virtual function of an interface shared by all three -- depending on the context and the similarity of the functions and the data they operate on. Neither philosophy is more correct than the other, but OOP makes it a lot easier to make massive, consistant, ubiquitous GUI applications.
The other point is, why are we waiting for the JVM to start up before we can do anything? Shouldn't it already be loaded and shouldn't it be trivial to load our program with a new classloader or as a new thread on the current JVM? After all, that's how Tomcat and other servlet engines work. Why is there no applet engine? I mean, once the JVM or CLI becomes the operating system...like it did with PocketLinux...
Hey freaks: now you're ju
The central debate here is about how to best use F/OSS development resources (people and code). The assumption seems to be that everyone who cares about F/OSS should come together on a single strategy for dealing with Microsoft. But a monoculture within the F/OSS community is exactly what we're fighting against in Microsoft! Must we become the enemy to defeat the enemy?
Most F/OSS developers want to see GNU/Linux succeed in the sense of becoming a widespread desktop alternative. Those who bother thinking about why they want this are most likely to come up with a fundamental reason: choice.
Survival for GNU/Linux is a question of the niches it is able to successfully occupy. For the moment these niches include the desktops of F/OSS developers and Internet server farms. The problem with the general user desktop niche is first that it's not really a niche per se, but more importantly, it's completely dominated by Microsoft. GNU/Linux is like a small mammal running around near the end of the age of dinosaurs. What it has going for it is adaptibility. Rather than give up adaptibility and become just another dinosaur, GNU/Linux needs to find another way to occupy the general desktop niche.
What I'd like to suggest is that the desktop niche really needs to be bifurcated in such a way that GNU/Linux can survive there as a small mammal, without needing to become a dinosaur. That is, it needs a place on the desktop where it can run without necessarily displacing Windows. One way this could be done is though a Windows port of User Mode Linux, but that's not really going far enough in my opinion.
What is really needed is an OSS virtual machine monitor (VMM) for PCs, under control of which both Windows and GNU/Linux (and any other OS!) could run separately and equally. Vmware shows what this might look like, but with Vmware the host operating system runs along side the VMM rather than on top of it. It sort of achieves "separate" but not "equal".
The problem with current approaches to PC VMMs is that they suffer from certain architectural limitations in virtualizing the CPU. These limitations probably could have been eliminated several hardware generations ago, were it not for the unholy alliance of Microsoft and Intel. But there is some hope that that alliance could be broken, if AMD would implement virtualizability in its CPUs and/or IBM would apply carrots and sticks to Intel on behalf of GNU/Linux.
The ultimate goal is freedom to innovate from the lowest levels of software on up. This can only be truely achieved by a complete OSS platform, as access to the source is what enables the kind of innovation that does not require reinventing the wheel when something at a lower level doesn't work the way you want it to. On the other hand, some F/OSS developers may be perfectly happy developing on top of Windows or some OS-independent application platform. Indeed, there's no reason to believe that .NET isn't "good enough" for some of them.