Slashdot Mirror
Sneaking Open Source Software Through the Front Door
the_1000th_Monkey writes "LWN daily pointed out this new project today. It aims to be 'a compilation of high quality open source software (OSS) [that] will be made
available as a CD distribution in order to help promote OSS to users of
Windows and MacOS.' There are hopes that this would make it easier to encourage universities, OEMS, and your parents/friends to take advantage of this software and eventually bring them over to a completely free system on their own time. Help for suggestions/discussion is being sought." Newsforge is carrying a slightly more in-depth look at this project. Anyone care to design some attractive, downloadable CD-graphic images?
Wasn't there something called cdrom.com that tried this idea sometime last decade? Did it succeed then?
---
Programming is like sex... Make one mistake and support it the rest of your life.
To answer the headline question. Sneaking a non-existant CD "Through the Front Door" is rather easy. I do it all the time.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
To get people to move to Linux, we must first get them to use alternatives to MS Office and other packaged software. Get them using the cross-platform software and then switching the underlying OS won't be as tough down the road.
The only thing that can stop Linux from eventually succeeding Windows on the desktop is either laws to prevent it from happening or not enforcing laws that will allow it to happen.
What we need is a mascot. Something furry and cute. Penguins are cute, Gnu's are not.
.... A OSS Otter? Cute furry, adorable frolicking otter. Okay artists, get to work.
How about
Agent K: A *person* is smart. People are dumb, stupid, panicky animals, and you know it.
What would be even better would be a cd that was full of open source software for windows. Like WinCVS, emacs, etc. If there isn't enough of it for a whole CD, write more!
Most open source software is for our open source operating systems. If we put more open source software out for windows, that is quality, people will use them, because they are free legally. If there are enough quality open source windows programs we can get to the point where people are using like 2 or 3 major open source programs a lot. Like how just about everybody uses WinAmp and AIM/ICQ.
What I'm trying to say, and very poorly at that, is we can make some open source programs for windows that will be very frequently used my numerous users. Then we can switch them to linux more easily, because it has all the programs they use every day and more. The biggest fright about switching to linux is that you have to use all new software for everything. All of your beloved programs are gone or different. People tend to find one piece of software for each task and stick with it.
The GeekNights podcast is going strong. Listen!
After all, open source software has been available for years on Linux ISOs...a lot of distros will install on a normal FAT32 filesystem.
The idea proposed is far different from a Free OS that will install on a FAT32 filesystem. It is modestly an introduction to the great big world of free software. Bringing a collection of Free Software that runs on thier existing windows system will allow for a smooth transition and eventual disconnect from the Borg that so many people could benifit from.
It is true that M$ has made a near science of useability and has made software that a well trained monkey could use. However, the Free Software distrubuted on this CD also has many of the "help systems" and useability features that the M$ bloatware has levreged to gain so much of the market share. There is no reason why any open-minded person would not swithch, or at least try out the software on this CD. They may even find that they can get just as much done for about $600 less than they could with thier M$ alternatives.
I've tried a number of times in the past to get my parents to switch over to Linux and OSS, and I used to recommend it to just about everybody. I always told people how easy it was to et up and use. This was a few years ago. Needless to say, I've learned a lot since then - the few people I've convinced to try it were never terribly happy, and after a week or so desperately wanted to switch back. I've learned my lesson, and now would never really recommend Linux or OSS to anybody unless they express an interest in learning more.
People who learned to use Windows and MacOS expect things from their software that OSS programmers have yet to really give them. What it boils down to is a polished product. I'll admit - I think the KOffice suite, StarOffice 6, The Gimp, and many other OSS projects are incredible. They're complete, relatively bug free, and give me all of the features that I want. But when compared with similar software on Windows and MacOS, most people find the OSS stuff just feels klunky. Most OSS software (let me stress the most - certainly not all OSS software is like this) just isn't as polished-looking to the degree that a lot of common Windows and MacOS programs are.
It's not so much about stability for the typical user. Sure, they want stability. They also want something that's intuitive, compatible with what their friends and coworkers use, and looks clean. They want software with a very adequate and easy to use help system, for when they get stuck.
Also, they don't want something with the exact same features as what they're currently using. After all, why switch unless you'll also be going to something better? The better the software can meld into how they currently do things, the more likely they'll switch.
In other words: yes the software needs to look pretty. Yes, it needs to be functional. Yes, it should be relatively bug free. But it also needs to integrate just as well or even better than what they're using now. It can't just be a functional replacement - it really needs to be something different and offer something remarkably better than current solutions. Finally, price isn't as much of a concern as most people think. That's not enough to make people switch from something the way they do things now, otherwise we'd have already seen the mass migration away from Windows and other closed-source, proprietary products.
--
Welcome to the land of the easily amused...
I think that this might work as a good "gateway drug" into the world of Linux for most users. I don't think that people are opposed to the idea of OSS, I think that they are just uncomfortable with switching their OS. If they kinda get a taste of OSS in an enviornment with which they are familiar, it might make them a bit more comfortable moving to Linux. I've always thought that there was a need for more Win32-based open-source projects.
Replacing MS Office is, in my opinion, the Holy Grail of open source target product replacements. What a product needs to do to compete with Office is:
If the above were true, I believe the product might succeed in becoming more widely used and supported.
In the mean time, a free version of an Office replacement might make it in the door and onto my desktop if it was good, but it would take the above criteria for me to roll it out to all my end users.
I do, however, commend this effort. The open source community needs some help putting it's best foot forward to be seen. There are some pretty darn good open source projects, but there is also a lot of noise that makes open source stuff look like free junk.
I would say the killer app for getting OSS into Windows boxen would have to be either rpm or apt-get.
If we could make installing software easy, painless, and reliable, we would have an open-source competitor to the Windows Installer. Just think: how many small apps use a full-blown InstallShield or Wise wizard when all they really need is to copy a couple of .EXEs and set up some shortcuts? But how many technophobic users would never install software by reading the README, unzipping the files, and putting them in the proper places?
Once you have a back-end like rpm or apt, all you need is a one-click graphical front-end that launches it. If installing the program all happens automagically, the users will think it is a Good Thing(TM). They will notice how they don't have to click through a gazilion stupid steps like ""C:\Program Files\Company-Name\Product-Name" does not exist. Do you want to create it?"
And then they ate Sir Billy's minstrels. And there was much rejoicing.
"Anything is better than IE, and you can quote me on that." -- Wil Wheaton.
The site: www.gnusoftware.com, which is nothing officially to do with GNU, is a collection of links to open source Windows projects. I've already used it to compile some pretty useful CDs to give away to people.
This sounds like a good idea as long as they keep it to best-of-breed products. One of the things I find annoying (as well as great) about Linux distros is the sheer number of applications avaialable. I would rather the distro only gave me 1 top quality CD palyer installed to start with and 1 browser, one office suite, etc. Then later if I wanted to allow me to try others.
Hopefully for this Windows CD they can stick with just a few top qualtiy products.
/b
[Please type your sig here.]
Looking at their forums, a lot of the people who contribute (to the forums, at least...) seem to be primarily Linux people who may use Windows a bit. As a result there's a load of people advocating crazy things like putting vim or emacs on the CD... Even programs like the GIMP are going to feel odd to most Windows users with their appalling X-style load/save dialogs...
Simple reason
e cs
/usr/lib/gcc-lib/i686-pc-cygwin/2.95.3-5/spec s
/cygdrive/c/devkitadv/bin/../lib/gcc-lib/arm- agb-elf/3.0.2/specs ../gcc-3.0.2/configure --prefix=/devkitadv --build=i686-pc-cygwin --host=i686-pc-cygwin --target=arm-agb-elf --e / --enable-multilib --enabl
#1
Reading specs from C:/mingw32/bin/../lib/gcc-lib/mingw32/2.95.3-6/sp
gcc version 2.95.3-6 (mingw special)
#2
Reading specs from
gcc version 2.95.3-5 (cygwin special)
#3
Reading specs from
Configured with:
-cpu=arm7tdmi --without-local-prefix --with-newlib --with-headers=../newlib-1.9.0/newlib/libc/includ
terwork --enable-languages=c++ --enable-targets=arm-elf,arm-coff,arm-aout --disable-win32-registry --disable-threads -v
Thread model: single
gcc version 3.0.2 (DevKit-Advance)
etc...
In case they didn't notice OSS is not really stagnant software. Putting it on a CD will be kinda useless because the software will be outdated in a month or two at most. Sure I guess putting only full whole [e.g. v1, v2, etc...] releases on it would be nice but just grabbing any old piece of OSS is a bad idea.
Tom
Someday, I'll have a real sig.
But let me reiterate what others have said: it must be EASY to use. No editing text to config. No compiling. No weird choices between tools nobody outside the free software world has ever heard of. Just a small, coherent set of useful applications that just happen to be free. Do that and you'll be amazed at how many people use and distribute this.
sulli
RTFJ.
"...eventually bring them over to a completely free system on their own time."
...gosh, had to be at least 3 years ago. I installed it...went "hmm" for a few weeks.. and then wiped it. A couple of weeks ago I thought it was high time to give it another look. Why not? I had listened to Linux users in these forums proclaiming how good this-or-that distribution has gotten that I kind of assumed that the operating system really was ready for prime time.
That implies running a "free" operating system, probably Linux.
"...your parents/friends..."
In my world this group would include people who aren't computer professionals and who's machines aren't otherwise professionally managed.
Having established all that... You're kidding me, right?
The last time I had run linux was
Uh...no.
I managed to get Yellow Dog 2.2 installed on my Titanium PowerBook, sure. But after having done it I was stunned at how...well, rudimentary the installer was. In spite of endless "the installer really has gotten good now" comments. Well, if by "good" you mean "nowhere near the usability of commercial desktop operating systems" then yeah I'll agree with that.
I managed to get my wireless networking going. I found out how to do a few other things. But the main thing I discovered is that Linux hasn't really gotten usable enough for novices. Somehow I expected more.
Frankly I don't care if some Linux zealot mods me down or lables my comment "troll" or "flamebait." While you're doing it, bear this in mind - I'm doing this for you. The Linux community really needs to take seriously the idea that a novice needs to be able to install, configure, troubleshoot, and maintain the thing without endless arcane documentation or professional help. If your fellow Linux users/developers won't tell you, I will.
Linux will never be more than a server OS and a geek toy until / unless the usability radically changes.
You like your Macintosh better than me, don't you Dave? Dave? Can you hear me Dave?
Anyone who needs those utilities already knows how to get them.People who use Windows or MacOS over Linux do it because the GUI handles all the details for them, not because they want to manage it all from the command line.
Take the GIMP. Quite powerful plugins. But nothing like the easy-to-use with presets plugins you get on Photoshop. Every time I see a GIMP plugin, I think "this is neat, but a non-techie won't go for it".
Take edge-detection. AFAIK, not a big deal in Photoshop. There are *three* different plugins implemented for gimp, each named after the algorithm they implement. This is cool if you're into image processing, not image editing.
If you run out and download plugins you can get cool stuff, too. The problem is that downloading, compiling, and installing plugins is not going to work for new users. And most of those plugins aren't oriented toward design sorts either -- more towards image processing engineers.
Let me give an example. A tasty-sounding plugin for the GIMP is Artistic->Oilify. Oil painting, cool! In Photoshop, if you run something like this, you'll get a little window with a preview, a bunch of bundled presets named "big gloppy brush", "Van Gogh", etc. In the GIMP, you get a frame called "Parameter Settings" containing a checkbox called "Use Intensity Algorithm" and a slider entitled "Mask Size". Now, maybe it's just me, but I doubt anyone but the original coder k(or someone that's looked at the code, or is familiar with this family of image processing algorithms) knows what the "use Intensity Algorithm" does. There's no preview, so no easy way to check. Mask Size doesn't mean anything to a non-coder.
The GIMP (1.2.3 ximian) still doesn't bundle even have a good, preset-capable, previewing drop-shadow plugin. This is something that people want, and usually they don't want to still run out and create another layer, fill the selection, gaussian blur and then offset the new layer. It's work, and the 1% of the time they want some weird effects in their drop shadow, they can do things the hard way.
That means you shouldn't have a "alpha squared" value slider, you should have a "hairyness slider".
And this is the GIMP, which is billed as just about the most consumer-oriented app on Linux.
AbiWord is probably one of the closest apps here to what I'm talking about. Any word processor user will be familiar with most of the options.
Every feature has to be documented, tooltips be included, etc. if people really want to try to take over the commercial app market.
Something like Apple Guide or Windows Help needs to be implemented. Tooltips should be implemented more than once in a blue moon.
Now, I'm not asking for anything -- I'm happily using and hacking on the software out there, and it works nicely for me. But if the intent is to go after the commercial apps market, then a few areas need to be addressed.
May we never see th
... is not to hand out CD's full of it to your friends and neighbors who aren't really interested in the first place. The most effective way would be to convince the businesses that employ these people to use it. People want to use the same OS and software at home as they do at work. Remember that most people were more or less forced to learn Windows and MS Office for their job, and have no interest in learning new software when they are already familiar with Windows (and MS Office, IE, etc...). If they were required to use Linux at work, they would be much more likely to try it at home. Familiarity is the key. So try sending your OSS CD to local businesses, and be sure to stress that all the software contained therein is completely FREE... even for business use.
------------------------- Thus Spoke the Mad Monkey
I'm discouraged that their list of "contributions from the community" doesn't include anything at all about decent documentation. The fact that the first item is "cool graphics" isn't very promising, either.
Accurate, well-written, and current documentation is absolutely vital. They apparently plan to link to "full online documentation", which are probably the cobbled-together FAQs and HOWTOs that are already available, and that's not likely to be adequate.
I never accepted Corel because they were a hybrid free/proprietary system. I love Linux, BSD, and GNU. I love RedHat and Debian, and would probably be happy with most of the other distros. I know there are flamewars, but I don't usually see them from the people I know in the community. (I usually browse slashdot at 4 or 5, though.)
The only war I'm really interested in is making sure they we are opposed to all proprietary software, not just MS. I could get along without MS but still be trapped in proprietary software from Apple or Sun. Plenty of people seem to think that attitude is petty, obstructive, or anti-business, but that's the one thing I'm passionate about.
My goal is not to get Linux accepted or to crush Microsoft. It is to get freedom and copyleft accepted. So, we're pointed in the same general direction, but at times you'll find me and people like me aren't focused in directly on the same things you are.
Secession is the right of all sentient beings.
Not a good difference.
The crowd this is aimed at would find that very confusing and aggressive. Installing a Windows binary by double-clicking is high-tech adventure for this crowd. Thousands of huge files of gibberish which Windows doesn't know how to open[1] are goind to look really bad to this crowd. It would seem aggressive because if they noticed it, they would assume they would have to do something about it. Why else would all that stuff be presented, if not because they must do something?
The folks who are susceptible to this don;t know what source code means, and they treasure their ignorance. Anything which threatens them with learning will cause them to run screaming. They don't want to learn.
.h and .c files in windows? Would you like to buy the Brooklyn bridge?
[1] Do you really think that some one with no compiler is going to have file associations set for
I managed to get Yellow Dog 2.2 [yellowdoglinux.com] installed on my Titanium PowerBook, sure. But after having done it I was stunned at how...well, rudimentary the installer was.
What do these two things have to do with each other? ``Ready for prime time'' and ``quality of installer''? Nothing. Granny can't install Linux, Granny can't install Solaris, and Granny can't install Windows. Despite the big advantage Apple has from their iron-clad control of everything which runs OSX, I suspect that Granny hasn't a prayer of installing OSX, either. There exists no OS which meets your criterion: `` ... that a novice needs to be able to install, configure, troubleshoot, and maintain the thing without endless arcane documentation or professional help.''
Toasters and refridgerators work that way, but complicated things like cars and computers don't. Do you think that cars aren't ready for the mass market? Once a knowledgable person sets up a Unix OS, it will run and run and run, for years. The user can't easily screw it up. That's ready for prime-time. Windows isn't, and may never be. Thanks to things like USB and Kudzu, Linux is pretty near there, once installed.
Have you tried a Windows installer lately? Some of them have pretty graphics, but their hardware detection and included drivers are way behind linux on Ix86. Installing Windows is HARD. Solaris is HARD too. Linux is a good deal easier to get running on intel hardware than those others. By the way, graphics doesn't make for an easy install. Good hardware detection and automated selection of the right drivers does.
You say that a big problem is getting the stuff installed...
I wonder if Gateway would want to promote this as a branch-off from their free music initiative. They already ship computers bogged down with other stuff, might as well put something useful on there too.
That could be a big deal for them. This software would add 'value' (as in functional or sales value, even though the software itself is 'free') to their products without costing them much(I say 'not much' because installing that on each computer will cost them).
"Never, never suspect the dreams within the dreams of dreaming children." ~The Amazon Quartet
When it comes to software with an inelastic demand, no one cares if it is open source or not. What businesses care about is whether it works and works well. If something was free but didn't work, or was a sub-standard product, no one would use it who could afford to do otherwise. A product that has a free price tag but hurts your productivity isn't exactly a good deal compared to a product with even a hefty price tag when that product helps you get your job done more efficiently. This is why companies like Oracle and SAS can charge as much as they do, the companies that buy their products SAVE or even MAKE money in the long run.
As for the political/religious free as in speech aspects of open source, you might as well be speaking swahili because most business brains won't understand what you're talking about and those few that do won't care.
If the idea behind this project is what I think it is, to make converts out of people so that they will shun commercial products in favor of free software irregardless of the quality of the latter, then the people behind this project are in for a nasty suprise: Non-hackers are indifferent to the open source movement. Some are going to be downright hostile in fact due to the way that some open source "advocates" behave.
What this project needs to do is promote the compilation based upon the quality of the software it provides, not on the basis that it is free. This will serve to educate the public that open source software can be just as good or better than commercial offerings. When this is accomplished open source will be able to compete with commercial offerings on an equal footing. It won't have a stigma attached to it and whether it is chosen or not will be entirely based upon quality and its suitability for a particular purpose.
I've been using Linux since 1995. I like the fact that it is free as in speech and in beer, but that isn't the reason I choose it over Windows. I use it because it is more powerful, more stable, and more flexible. If it were kludgey, flaky, or unstable then I'd do little more than play with it. I certainly wouldn't use it in a commercial setting where downtime equals dollars down the drain. This is the burden that all software must carry regardless of how much it costs or how accessible the source code for it is.
Lee
Muslim community leaders warn of backlash from tomorrow morning's terrorist attack.
We don't want to run a different OS on the system. We just want to introduce the people to the available OSS software that will run on Windows while they still run their everyday programs (MSWord).