I'm in the process of switching from vim to emacs right now. It's frustrating, although viper-mode takes a lot of the pain away, and to be honest I find it a really nice combination to have the vi editing modality readily available within the powerful framework of emacs. Remapping caps lock to Control has helped enormously.
The most notable thing is that emacs encourages you to live in it much more than vim. With vim, I would navigate around in the shell, open up a file in vim, edit, close, repeat. With emacs I'm finding that I spend more time just switching buffers instead of running back to the shell because emacs doesn't simply quit when I kill all my open files. There's always more buffers, so the app stays open and I tend to remain there.
I'm still in the process of getting up to speed with basic editing things, although I've found that the way it manages buffers and what it calls windows (what the rest of the world calls frames these days) is quite easy and quick to use, and I found myself able to navigate around in emacs fairly quickly. All in all, I'm not 100% sure I'll stick with it, but I'm happy so far and I know that I'm still getting over the hump to where I'll be truly productive with it.
If you are anything like me, that is, someone who loves the speed and "feel" of vim, but also enjoys the buffers of emacs. You might want to check out vile (vi like emacs). [http://invisible-island.net/vile/]
>With vim, I would navigate around in the shell, open up a file in vim, edit, close, repeat.
I do that to. I open up an `eshell`, navigate around. Then I edit a file with `vi filename.py`. That works best if you have `alias vi find-file $1` in your eshell aliases.
Pick one that's generally associated with web stuff. That's pretty much Ruby, Python, Perl, Java, and PHP these days. Take a look at all of them on the surface level to see what's interesting to you, and just pick one and go.
Given that you already know C and C++, I'd learn Perl, Python, or Ruby, since they're a very natural extension of C and C++ and can be used very easily in problem domains where you wouldn't want to use C or C++. Learning more than one of these may prove useful if you want to hack on a variety of different open source codebases, so whichever one you pick first isn't so important as picking one and having it in your toolbox.
The comments show that the article author presents a false dichotomy. All the data collection and analysis depends on models. Then from that analysis you get new models and do new analyses. This is how science works, whether you have tons and tons of data or not.
Nothing has failed yet. You act like the project is over and done with. It's not. Linux has conquered pretty much every single space but the desktop. This is a huge success, especially when ten years ago it was exciting to read about any company deploying Linux publicly in any sort of way. Now it's expected.
As for the desktop, the work isn't finished, and it'll take a very long time to get to the point where it's a real contender. It's not a QA issue. The Linux desktop of today is actually of very high quality. It's that it's a very difficult space to compete in.
Most folks don't realize, but desktop Linux has surpassed Windows and OSX in convenience, features and usability. Sure it still lacks some critical desktop applications and high quality UI design to truly shine. I've been alternating between Ubuntu and OSX for quite some time now, and dumbness of OSX (and it's visual polish) amaze me to this day: brand-new "naked" Leopard needs quite a few add-ons (sometimes commercial) to get on par with even (!) Gnome.
Lets face it: conquering consumer market is impossible without millions spent on "Mac vs PC" commercials and I wouldn't expect Linux to ever get there, unless PC manufacturers start paying for it. Linux gets a nice chance to be a major player whenever DELL/HP/whoever develops a nice line of Linux-only machines with a catchy name and distinguished set of applications, accompanied by a massive PR-campaign, kind of like Eee PC on steroids.
With the exception of a few distro's (Ubuntu and Sabayon are the only ones that spring to mind), how many ship with Fusion enabled? Let alone working graphics drivers needed for Fusion to run? It's great that one or two distro's can do this, but I thought the strength (and bane) of Linux was supposed to be that there's so much choice and ability to customize things. That you're not locked into a single distro; yet if you want things to work right away, you're still locked into a limited number of choices.
You mention that linux needs a "…catchy name and distinguished set of applications…" before it can gain popularity or widespread usage. How did Linux manage to surpass other OS's if it doesn't have the same/similar features that people use?
I would like to have a healthy discussion here but your post seems to be a set of randomly collected issues without a core point holding them together. I could respond with a similarly random bag of rants regarding OSX or Windows or anything else, really.
But I won't. The original statement was that "Desktop Linux Failed". I objected and offered a way to "Succeed": IBM/Lenovo could have taken their existing ThinkPad line with its 100% hardware compatibility, put a prettier face on a hundred or so selected Gnome apps, give it a cool name and spend $50 million promoting them on TV, exactly what Apple did to BSD.
Ah, sorry about that. Didn't mean to be randomly collected, it was more of a spur of the moment post. It's not worded in the best way possible, but there is a core point in there.
The basic point of the first part of my post was supposed to be that very few distros contain all the features that make the distro equal to OS X or even Windows. Because of this, it's inaccurate to say that Linux as a whole is equal.
The second part of my post was a reply to a statement you said. It makes a lot more sense if read separately instead of part of the same post.
To reply to what you said in response to me (Pretend this is the start of a separate post for more coherence if you like): What reason is there for IBM/Lenovo to do such a thing? Apple had a clear reason for buying Next; the failure of Copland. But how would taking Linux, developing a new UI, and rebranding everything else help IBM/Lenovo?
"With the exception of a few distro's (Ubuntu and Sabayon are the only ones that spring to mind), how many ship with Fusion enabled?"
How many people really need Fusion? The vast majority of distros are niche designed to showcase a particular innovation or for a particular application, which may have nothing to do with general desktop use. Many others are designed for 'enterprise' applications and also have no need of Fusion. Yet more are designed for power users, who may or may not want Fusion, and certainly know how to install it if they do. only a few distros are targeted at casual users and of them, only a couple think that Fusion is the proper solution.As you must understand, linux about flexibility and just because a particular distribution doesn't pack the features that you think it should have doesn't mean that it doesn't suit its target audience very well.
"How did Linux manage to surpass other OS's if it doesn't have the same/similar features that people use?"
Exactly. Linux is about flexibility. However, for the end user (and unless I read this article completely wrong, the article discusses software aimed at these people) to get something set up exactly how they want it, Linux doesn't make this an easy task to do quite yet. Like you said yourself, there's distro's aimed at different tasks.
Ideally (to me), some distro could have website that could autodetect hardware and have a bunch of {check, drop down, radial} boxes for software to put on it. (Want Fusion? OpenOffice? Gnome or KDE or XFCE or so on? How much information to see? Everything/Quick blurb/Errors only/Whatever else? Hell, if its for the end user, let them upload their own desktop background to use right away. Little things are important too) The site would then customize the distro and create the iso as needed for the individual.
That would in itself go a long way to creating a usable customizable OS that does what people want instead of trying to guess what they want and create yet another variation within the sea of distros. But, such a thing hasn't happened yet for a variety of reasons.
Just because I can customize a {Gentoo, Slackware, Debian, Red Hat, etc} install to be exactly how I need it, doesn't mean my neighbor who "heard about this lunix thing" can do the same.
"Different target market."
Which one are you discussing? And is it the same one as I am? I'm talking about end users.
The problem is that you (and the article) seem to define the 'end user' too narrowly. Most distros weren't meant for that particular person and shouldn't be pushed on them.
The OP criticizes open source as a 'failure' because it doesn't satisfy a demographic for which it was never meant.
'End users' as you describe them, don't need Linux, they have Windows, and for them, it works just fine.
If the "end user" isn't the neighbor who buys/uses software, then what is it?
Because if you define "end user" as businesses, then they're going to come in contact with Linux int he office. If "end user" is defined as "power users" then Linux is doing just fine and every "This is the year of Linux" or "My technophile girlfriend uses Ubuntu!" article is wrong off the bat.
"If the "end user" isn't the neighbor who buys/uses software, then what is it?"
Whoever you wrote it for. In open source, it is usually yourself and people like you. Sometimes its businesses, or non-computer people, but usually its 'hackers.'
Gentoo doesn't come with Fusion because it's target user base is made of technophiles who like to tinker. Red Hat doesn't have it because it is targeted at business users who use it to power servers and backends. Ubuntu targets 'ordinary users' and does come with Fusion. It's really that simple.
"Ideally (to me), some distro could have website that could autodetect hardware and have a bunch of {check, drop down, radial} boxes for software to put on it. (Want Fusion? OpenOffice? Gnome or KDE or XFCE or so on? How much information to see? Everything/Quick blurb/Errors only/Whatever else? Hell, if its for the end user, let them upload their own desktop background to use right away. Little things are important too) The site would then customize the distro and create the iso as needed for the individual."
That would be a cool distro but I don't think that it's feasible. Besides, part of the fun of Linux is exploring what's available. Ubuntu comes with a large repository and it makes it easy to explore all the different applications that are available. Trying different things out was one of the things that always kept me on Linux, it was a lot more fun than the Windows model, never mind that I couldn't use Photoshop.
Thanks for the clarification on your definition of "end user" - we were just looking at things from different perspectives I guess.
As for the feasibility of the idea? Its not feasible for me to do - beyond my knowledge. But I don't see why someone can't manage it at some point in the future if they wanted to. But that could just be because I thought of it and no one likes to think that they've thought of a bad/impossible to implement idea.
I have better things to do than invest time in trying Linux again. Gnome may have features that Leopard doesn't have, but my subjective probability that Ubuntu just-works like my Mac does is very low, and I'm a geeky user.
This isn't about commercials. Millions were spent in server-space commercials and Linux still came out winner because it was a cheap good-enough solution. Large companies have given desktop Linux many opportunities to catch on because it's economically more profitable. The crummy QA of desktop distributions (server distributions also have crummy QA, it's just that the geeks who run them are able to surf through it) made desktop Linux fail.
The EEEPC (and other linux-based low-cost PCs) success is a positive indicator that Linux can make it. Sure, some "geeks" will install XP on EEEPC, but for the most part, people are using Linux on it.
I bet you haven't used any Linux in years. It's changed... alot.
It isn't such a hard space to compete in. There's a huge movement towards cheaper PCs. But anyway, for a while there was a huge gaping window when Linux was "cool" (the way OS X is now) and many many people tried it. The QA sucked by then.
I mean, I was a true believer -- I went all the way building a Gentoo desktop and whatnot. But none of the readymade distributions I tried (SuSE, Mandrake, Fedora, etc.) were up to the task of fueling a working environment where I don't have to worry about the working environment.
After a while, I just got a Mac and quit bothering with useless geekery. I went geeking into functional programming instead.
Did you get the news that the next batch of OLPC laptops will use Windows XP? They weren't bribed by Microsoft. It's not a matter of interface either, they'll keep the same interface.
It's actually a very difficult space to compete in. Having worked on major desktop-related infrastructure for a major linux distro, I've got real experience on this one. It's a decidedly non-trivial task.
Back when linux was "cool" (when was this? Late 90's?) only geeks tried it. Some stuck with it, others went to OSX, like yourself. Most people, during this time, managed to hear the word linux, but didn't really know what it was. Consequently, they didn't try it. This mythical time when everyone and their grandmother was trying Linux just didn't happen. But just because Free Software didn't work for you back then doesn't mean it failed.
Mac OS, by my standards, failed miserably. Totally. Utterly. So I went to Linux. This was pre-OSX, when Apple was a company that had totally lost its way and was getting demolished by Microsoft. Pretty much everyone said that the Macintosh and Apple had failed and their days were numbered. Now things are different because Apple didn't stop. They never really failed you see, they just kept going. The Free Software community is doing much the same, and I'd be surprised if we're in the same place in another ten years.
Just because something is not meeting your specific needs right now doesn't mean it's failed. Given the enormous progress I've seen over the past ten or so years I think it's fair to say quite the opposite. You actually have a choice to run Linux today where that wasn't really an option a decade ago. If that's failure, then you have a strange definition of failure.
Did you get the news that the next batch of OLPC laptops will use Windows XP? They weren't bribed by Microsoft.
Sure, but the governments OLPC needs to sell to in order to succeed, were bribed (and threatened, and lied to, and pretty much all of the other tricks in Microsofts 30 year history of FUD and scorched earth tactics).
And, of course, all but Negroponte were disgusted by the change of direction and left the project.
Sure, because helping kids in the third world is suddenly not important if it's not gonna help promoting your relgion, er, platform of choice.
Did everyone but Negroponte _really_ leave in disgust? Boy, the zealotry actually surprises me, and I spent six years watching Slashdot for several hours a day.
because helping kids in the third world is suddenly not important if it's not gonna help promoting your relgion
Now you're just being retarded and resorting to name-calling (see what I did there?). I write both Open Source and proprietary software, and I have no religion.
You've already shown yourself to have a deep misunderstanding of Open Source, and its community. You've also shown a willingness to erect straw men and hold up single individuals as representative of an extremely diverse group of people. And, you've shown a profound lack of comprehension of how far Open Source has worked its way into every aspect of human technical activity, and how it continues to grow.
Your article may have been interesting and mildly believable ten years ago (and I remember reading lots of similar arguments back then, and they were a little worrisome to those of us in the Open Source community), but today it just exhibits an extremely limited view of the technology industry and where Open Source fits within it (where==just about everywhere).
In short, you're not a credible witness, and you become aggressive when someone calls you on your weak arguments.
I think there was more to it than the switch of platforms. If I recall, Negroponte expressed a notion that he was more interested in selling the pcs than helping children.
I still don't understand the OLPC move. The whole _point_ of the things was to teach a man to fish, not to give him a fish.
From that basis everything (even the hardware) was open so that it could be studied and changed. It made perfect sense. The sellout doesn't seem to gain anything at all (as you say, it even uses the same interface) whilst giving up the fundamental principle of independence for the recipient.
OSX would fail completely without free software. BSD is the subsystem, their browser is based on konqueror's html engine, they rely on cups for printing, samba for file sharing with windows... the list goes on and on.
What's important about this isn't so much that free software didn't fail, but that modern computing could not and would not exist without free software. End of story. Mac classic would be all but dead now, and Windows would have "innovated" us all in to a ditch. Remember what Microsoft did to IE when Netscape wasn't a threat? It'd be the same story. Without free software, there wouldn't be a mozilla, an OSX, or anything like it. Modern computing absolutely requires free software in order to function.
Sure, but the major cost center for PCs is hardware, not software, at least not the operating system anyway. Major manufacturers are permitted to install Windows on their machines for only a small fraction of the retail price, so installing Linux instead doesn't save much money.
Either way, the main driver for open source isn't price, it's flexibility. Not everyone needs it, but for those who do, it's a huge boon.
> the major cost center for PCs is hardware, not software
Microsoft has increased the cost of Windows licences while the cost of hardware components has decreased. The result is that the "free" copy of Windows that comes with pre-built PCs takes an increasing slice of the costs. If a hardware manufacturer can sidestep the OS cost while providing its customers with a functional computer then it gives them a huge competitive advantage.
Is there any word on whether other browsers will be implementing this in the future? It's what's keeping me from using canvas today, and partially what's keeping me from writing a certain app I've been thinking about for a little while.
I've tested some of my SVG demos with FF 3 and they run much faster than with FF 2 (I think that's due to Cairo being integrated into Firefox 3). Opera and Safari are very fast compared to FF 2, too.
As for feature completeness I agree with you. The Firefox SVG team is very small and not funded by the Mozilla corporation (at least that was the case last year when I talked to one of its members) - the SVG standard, on the other hand, is HUGE.
I'm using it for a simple web app, testing in FF2/FF3/Safari, seems to work fine... there are some annoying bugs in the implementation (the kind that have been open for years), and inconsistencies between Safari and FF, but nothing I couldn't work around so far.
We started by using graphviz to generate svg, but then couldn't find any decent in-line svg viewers. Do you know of any non-java applet inline svg viewers?
The most notable thing is that emacs encourages you to live in it much more than vim. With vim, I would navigate around in the shell, open up a file in vim, edit, close, repeat. With emacs I'm finding that I spend more time just switching buffers instead of running back to the shell because emacs doesn't simply quit when I kill all my open files. There's always more buffers, so the app stays open and I tend to remain there.
I'm still in the process of getting up to speed with basic editing things, although I've found that the way it manages buffers and what it calls windows (what the rest of the world calls frames these days) is quite easy and quick to use, and I found myself able to navigate around in emacs fairly quickly. All in all, I'm not 100% sure I'll stick with it, but I'm happy so far and I know that I'm still getting over the hump to where I'll be truly productive with it.