LXer compares Ubuntu, Mac OSX, and Windows XP. “All three desktop operating systems have admirable qualities. Each has some weaknesses. Attending a recent User Group Fair, I had another chance to see them at work. Having used and programmed on each platform provides some unbiased insights.”
Ubuntu, Macintosh, Windows XP
About The Author
Follow me on Twitter @thomholwerda
2006-03-21 12:04 pmBryanFeeney
Not to mention that he manages to use Ubuntu “for 12 or more hours a day” and still find time to investigate Windows and Mac OS X.
Then there’s this
“Microsoft’s development tools work strictly on Windows and do not allow for interoperability amongst desktops such as the Mac, Linux or systems running non-Intel based processors.”
Of course, this is true of XCode on the Mac, and KDevelop on Linux, but is not mentioned as a demerit for either. Secondly, Visual C# does, arguably, allow for interoperability across desktops such as Windows and Linux seeing as it can create .Net runtimes that can be executed by Mono. You may need to use the GTK# libraries to do this, but Microsoft’s development tools don’t prohibit using third-party libraries, in fact they make it quite easy.
As for users paying more for “Apple’s performance”, the Powerbooks and iBooks were, until recently, drastically underpowered compared to their Intel-based counterparts. Likewise the G5, with it’s weak integer performance, is quite underpowered for certain tasks in comparison to Intel. One prime example is compilation, which makes it a lot less friendly for developers (the Intel iMacs out-perform the dual-core G5s in this area).
“People who can afford Apple products have a special devotion to the Mac and OS X” is another fun claim: people buy Apple products for an easy, all-in-one hardware and software (iLife, iWork and the media browser) system which is easy to use, and which looks good.
Also I take issue with Ubuntu being good for everything. I’d say Redhat Enterprise is a better enterprise offerings, especially on the server, as it is well established, has a working support system, and is supported by all major ISVs, including heavy-hitters like Oracle. Ubuntu won’t release a supported enterprise distribution till June, until then it’s all smoke and mirrors.
All in all, a disappointing, biased (Linux = cool, Windows = teh sux, Mac OS X = teh bling) fluff-piece
Edited 2006-03-21 12:16
I see a very high level, very low information overview of three platforms, with an apparent primary purpose of offending no one (as if!)
Given the credentials of the writer, I expected useful information. Instead, it reads sort of like Consumer Reports, minus the hilariously clueless recommendations CR is known for. Anyone in the market for this article probably still couldn’t understand it (and certainly doesn’t read OSNews.)
What was I supposed to get out of it?
You know I really like OSnews, but this was the worst article I have ever read.
Post your favorite parts! I liked:
>> I have a personal bias for Free Software, but I’m wearing my certified analyst hat in an effort to remain objective.
I have never met anyone who is totally unbiased:
“I use Ubuntu for 12 or more hours a day.”
Unbiase life: 8 hours sleep + 5.33 Linux + 5.33 Mac + 5.33 Windows.
That was not a good article, i was expecting this really detail lets compare each system good and bad. Instead we get a non-offending general skimpy light overview with a neutral conclusion with seems articfical. Everyone has an objective and a conclusion. True no-one OS is perfect. But for a guy who uses Linux 12 hours aday we think we know where his true bias lies. Also is it me, or does this guy spend half the time justifying why he is qualified to write the article then just stops with a light conclusion…
Edited 2006-03-20 18:28
2006-03-20 9:07 pmnickbotulism
amen. red flags went off as soon as i read him declaring his insights as “unbiased”. even if the author was using all OS’s an equal amount of time, it wouldn’t magically remove his preferences. what’s more important when reading any opinion is knowing what their biases are.
That’s all this is … an opinion article. I have opinions, you have opinions, Thom has opinions — and they probably all differ. No single opinion is more correct than any other one when it comes to what OS you prefer.
In my experience, for instance, I find that Ubuntu runs slower on the same level of hardware than OS X/Windows XP do. He says the opposite. Uh oh.
Is OSNews going to publish *any* opinion piece written by some guy who’s used a lot of OSes in his time? If so, I’m sure half of us here would meet those same criteria — I’ve been playing with various OSes for most of my life.
2006-03-21 1:49 amarchiesteel
Yes, but you have a known anti-Linux bias, therefore one should take your claims with a grain of salt…
2006-03-21 7:05 amAnonymous
How is that different than having a pro-Linux bias?
Ooh, food for thought, isn’t it?
2006-03-21 5:48 pmarchiesteel
Being pro-Linux doesn’t mean you are anti-Windows. It means that you have a positive preference towards a particular OS. You have a negative attitude towards that OS, which means that you’ll see faults with it where there aren’t.
The author of the article stated that he had a pro-OSS bias, but managed to stay balanced in his evaluation. Of course, an anti-Linux poster won’t be able to accept this, because it does not fit in his agenda. However, a pro-Windows or pro-Mac OSX poster would not have problems with the article, because it highlighted positive aspects of those OSes as well.
In other words, only anti-[insert OS here] posters would have something to complain about this article. Pro-[insert OS here], on the other hand, will find it a good (if a bit light) article.
P.S. Before giving me “food for thought”, you’d first have to make a convincing argument. This hasn’t happened yet.
Edited 2006-03-21 17:50
2006-03-21 9:03 pmAnonymous
> Being pro-Linux doesn’t mean you are anti-Windows.
Are you sure you want to place money on that? 😉
> It means that you have a positive preference towards a particular OS. You have a negative attitude towards that OS, which means that you’ll see faults with it where there aren’t.
It can, but it can also mean that I’ll see faults with it where others are not willing to admit there are faults. With you, it’s very possible that you’ll see positive things about it where there aren’t any positives. The street goes both ways, buddy, because both points of view are equally misleading.
> The author of the article stated that he had a pro-OSS bias, but managed to stay balanced in his evaluation.
Way to contradict yourself in the same sentence. The very definition of a bias means that you will not be able to make a balanced evaluation. It was an opinion piece, and nothing more. My opinions differ from his, but I guess he’s correct and I’m wrong because he loves OSS and I don’t? Please.
2006-03-22 6:01 amarchiesteel
Are you sure you want to place money on that?
Yes. I am pro-Linux, but I’m not anti-Windows. Pay up.
It can, but it can also mean that I’ll see faults with it where others are not willing to admit there are faults.
That’s what you claim, but I’ve never seen you actually point real faults. Rather, you take small flaws and make mountains out of the molehills they are.
With you, it’s very possible that you’ll see positive things about it where there aren’t any positives.
If it’s positive for my needs, who are you to say that it’s not positive?
The street goes both ways, buddy
I’m not your buddy.
Way to contradict yourself in the same sentence. The very definition of a bias means that you will not be able to make a balanced evaluation.
No. I can be biased, but that in itself has no effect on the validity or not of the arguments I put forward. In rational debate, one analyzes and responds to the arguments, not the person who makes them.
Here, the man admitted his bias, but he presented his case in quite a neutral view. That’s because he didn’t have an “anti-[insert OS name here]” view, and therefore didn’t feel like he needed to attack anyone to make his case.
If you can’t see the difference between negative and positive criticism, then that explains a lot about your own social behavior.
2006-03-21 8:35 amormandj
I pretty much ignored the followup post to this one, as it seemed pretty irrelevant to the discussion. I would like to, however, discuss one point of your comment.
“In my experience, for instance, I find that Ubuntu runs slower on the same level of hardware than OS X/Windows XP do.”
I think this can be said for 99% of the desktop systems on earth, doing anything productive. People who make HW spent fortunes optimizing for windows. The end. What would you expect the end result to be?
It’s different in the server-world (having just come back from a 7 day knock down fight with a dozen win2003 servers.) Linux blows win2003 server out of the water for serving. Solaris buries them all imo.
It’s all about optimization. The same hardware is going to perform differently based on the controlling software. Windows is very well optimized as a desktop (well, one could argue not so well, but I think better than linux!) Unixes are more server-oriented (traditionally, this is slowly changing.) Once optimization shifts to desktop linux, you’ll see it start blowing away windows. That’s just how stuff works. Remember DOS? I had dos games, that run in win98 were slower even on better hardware. Now, roles have reversed. We’ll see how it goes for desktop linux, there is a lot of usability issues to get over first.
2006-03-21 9:16 pmAnonymous
I disagree with pretty much everything performance-related you’ve said. 😛
There’s no such thing as “optimizing hardware for Windows”, unless you’re talking about Direct3D-related silicon on video cards. The whole “OMG Windows is only faster because the hardware is optimized for Windows” is an old and tired, and as of yet unproven, argument.
Your comment about Linux blowing Windows Server 2003 out of the water for serving is also unfounded. Can I have actual specs and benchmarks? Do I know how you set the two systems up? Nope. Hearsay. I don’t want it, unless you can prove it.
Thus far, it’s been shown that IIS6 can serve more requests than a Linux box with Apache, but that’s all I can recall. I haven’t seen any conclusive and fair benchmarks between the two for I/O, DB, etc.
The only way I’ll see Linux blowing Windows out of the water in desktop performance is if X, the window managers, the toolkits, and the drivers are all fixed and optimized. Right now, GNOME and KDE feel like molasses compared to XP. I don’t see that changing with just “optimization”. There’s simply too much bloat in modern Linux distros.
2006-03-21 9:30 pmormandj
“There’s no such thing as “optimizing hardware for Windows”, unless you’re talking about Direct3D-related silicon on video cards. The whole “OMG Windows is only faster because the hardware is optimized for Windows” is an old and tired, and as of yet unproven, argument. ”
I was speaking about drivers/software. Again, lay off the semantic argument, I’m really dumbfounded if you didn’t understand what I meant. Sorry for not being clear enough – I assumed your innate knowledge would make it obvious what I was referring to. Assumption is the bane of sane existance I suppose.
“Your comment about Linux blowing Windows Server 2003 out of the water for serving is also unfounded. Can I have actual specs and benchmarks? Do I know how you set the two systems up? Nope. Hearsay. I don’t want it, unless you can prove it. ”
This has been beaten to death. A quick google should resolve this for you unless you just don’t *want* to accept the facts. You’ve never run both if you don’t know this already. I personally use Solaris, and I’ve adminned a ton of win2003 servers as well. I’m telling you, in every scenario I’ve been in, win2003 has constantly been out-performed. That’s my experience. I’m not writing a benchmark spec sheet for you, I’m just telling you my experience. You’re free to believe whatever you like. Google (and by that I mean (seemingly) most people) would tend to agree with me.
“Thus far, it’s been shown that IIS6 can serve more requests than a Linux box with Apache, but that’s all I can recall. I haven’t seen any conclusive and fair benchmarks between the two for I/O, DB, etc. ”
Uh, sure….. “Can I have actual specs and benchmarks? Do I know how you set the two systems up? Nope. Hearsay. I don’t want it, unless you can prove it. ” Have fun with your own words. In my experience (as the owner of a data center heh..) this is absolutely untrue. Maybe in some niche case, but in *my* real-world usage (and all of my customers) this is absolutely incorrect. I’m constantly migrating clients off windows to other solutions (be it solaris/linux/whatever fits the bill) – but running on the same hardware – and it is *consistantly* improving performance by large amounts to do the same task. Blame it on poor daemons on the windows side, or whatever – but the overall experience is better with the unix servers. Take it as you may, it’s my subjective observation (as well as my clients.) Going from only being able to process X amounts of queries/page requests/whatever to Y amount, where Y==10*X is proof enough for me. I don’t need to benchmark the hell out of it to prove anything to my client, nor myself. Again, feel free to disbelieve my experiences. (And don’t bother googling.)
“The only way I’ll see Linux blowing Windows out of the water in desktop performance is if X, the window managers, the toolkits, and the drivers are all fixed and optimized. Right now, GNOME and KDE feel like molasses compared to XP. I don’t see that changing with just “optimization”. There’s simply too much bloat in modern Linux distros.”
You just said the *same* thing I said before! One minute you say there is no such thing as optimizing for windows, now you say the opposite. Please don’t retort I was referring to optimizing HW for windows, I’m not an idiot. I also specifically spoke about software in my last paragraph. Try reading it this time.
“It’s all about optimization. The same hardware is going to perform differently based on the controlling software. Windows is very well optimized as a desktop (well, one could argue not so well, but I think better than linux!) Unixes are more server-oriented (traditionally, this is slowly changing.) Once optimization shifts to desktop linux, you’ll see it start blowing away windows. That’s just how stuff works. Remember DOS? I had dos games, that run in win98 were slower even on better hardware. Now, roles have reversed. We’ll see how it goes for desktop linux, there is a lot of usability issues to get over first. ”
C’mon bud, open your eyes and quit arguing blindly in the dark.
2006-03-21 12:11 pmBryanFeeney
No single opinion is more correct than any other one when it comes to what OS you prefer.
Yes it is. I really think this “every opinion is a beautiful little rainbow, to be cherished and loved” stance is utter nonsense. While you can respect someone’s right to give an opinion, and should be civil in responding to it, if an opinion is not backed by verifiable facts, is self-contradictory, or is uttered by someone lacking the sufficient knowledge or experience to analyse the subject matter, then it can be less correct, or valuable, than others without this weakness.
And lets not forget, the point of this article was not to state which OS the author preferred. It was, he claimed, an “unbiased” survey of the three most popular operating systems, carried out by himself with his “certified analyst hat” on. He attempted to offer an expert opinion, which I respect, but I don’t value his opinion, as it is inconsistently argued, not backed by facts (there is no currently supported Ubuntu distribution for the enterprise) and contains claims of highly dubious merit based more – it seems – on subjective personal opinion than surveys or investigation.
Edited 2006-03-21 12:12
2006-03-21 9:05 pmAnonymous
Ohhh don’t get me wrong — I’m not one of those people that thinks everyone deserves a respected opinion. You know that saying about opinions and a**holes? Yep.
Besides, there is no such thing as “personal opinion” when it contradicts reality.
It was an opinion piece, but nothing more. Like you said, it lacked facts and was inconsistently argued.
As one of the readers has already commented above, it’s just an opinion piece, but, nevertheless, throwing around meaningless descriptions like characterizing Ubuntu as a good “midlevel desktop”, after he just described all the comparable software in existence for GNU/Linux that make the end user experience similar to that of Windows, is confusing. He seems to like it, but never goes on to say what a “high level” or “premium desktop” is. Perhaps, I missed it. Did anyone get a sense of which currently available OS the author actually favors?
I think it was a good article because it said what needed to be said — nothing more, nothing less. And it comes from an system’s administrator, it seems, which gives credibility to his opinion to a certain extent; also, other system administrators probably will relate with this tone and ideas.
Overall, it’s a nice have for OSS.
Honestly, it was bad enough this POS article got on digg, but now its on slashdot and OSnews!?#!#@$%&%
Are my eyes deceiving me, are people really this stupid? To think this tripe shines any new light on this topic blows my mind.
The internet truly sucks.
” It lacks applications such as Photoshop, Framemaker, Pagemaker, Visio, Access, Quickbooks, legal DVD players and most importantly income tax preparation software. ”
Photoshop = you said it, GIMP
Visio = OO.org Draw
Access = OO.org Base
Quickbooks = runs under WINE fine! (you want native GnuCash, Quasar, SQL-Ledger, and AccPac) or for most OpenOffice.org Calc
a PDF converter = is built into OO.org, I prefer to save as PostScript though, which works from any app than can print
legal DVD players = VLC + libdvdcss + libdvdplay
tax preparation software = as you said the internet, or again, WINE
And he shoud have said “Premier”, but wait there is Cinelerra….
“Without those applications ported directly to Linux, Ubuntu remains a mid-level desktop.”
you might think so, my users don’t
2006-03-21 2:22 amormandj
GIMP != even close to phptoshop
Access != even close to OO.org Base
Quickbooks != any of those alternatives, and why run QB non-native under WINE?
PDF functionality really isn’t that great
I don’t see libdvdcss/libdvdplay being legal? Not in the USA anyways, but maybe I’m confused.
Tax prep software – why run non-native under WINE?
Cinelerra I can’t comment on.
I don’t know who your users are, but they aren’t professionals in the universe I live in. The gimp/PS thing takes the cake. Nobody in the gfx/art field would last more than half a nanosecond without laughing after hearing somebody compare GIMP and PS.
All in all, if you need most of those things you listed, you’re plain better off just running Windows, or writing/improving the linux software to the point it IS comparable. It’s not now.
2006-03-21 2:49 amTechGeek
why run non-native under WINE?
How about the fact that I dont have to worry about spyware and virii stealing my tax information while I am filing? Or reporting my account numbers to some thug in Russia? Of course, not all of us have $900 to shell out for photoshop. Gimp does 90% if not more for free. And whats wrong with pdf suuport? Works just fine on every version of linux I have used without having to download it ffrom Adobe. BTW there is a free version from Adobe for linux, so you cant say much about it. And yes, I am biased toward linux. But if you are going to be a ms fanboy, at least pick some things Windows is actually better at.
2006-03-21 5:54 amormandj
I don’t even own a MS-running machine. All my systems are mac/freebsd/solaris. Sorry to disappoint about being a fanboy.
I don’t know what your spyware/virii stealing your tax information comment was about, but that’s simply not the case. Every infested windows machine I’ve ever seen has been a user’s fault. The only exception to this was the whole blaster thing. Dumb users on any OS are going to have infested systems. It can happen on OSX, it can happen on linux, solaris, you name it. User-owned files can be infested to hell and back. It *appears* more severe in windows because of the default-admin privs – but who really cares about the OS – it’s the data that matters. And that’s what gets screwed, no matter what.
Uh, 90%? No way. I’m sure it’s more than capable for a basic user, but in no way is it 90% feature complete of PS, I won’t even get into usability.
PDF support – it’s been a while since I’ve attempted in linux, admittedly, so maybe it’s better now. It used to be I couldn’t even fill in PDF forms. I won’t attempt to knock your retort though, it very well may be possible support is improved.
Your reasoning/logic fails, I’m sorry.
2006-03-21 4:46 amchemical_scum
Access != even close to OO.org Base
Exactly you said it.
GIMP != even close to phptoshop
Perhaps you meant Photoshop which runs under Wine
I don’t see libdvdcss/libdvdplay being legal? Not in the USA anyways, but maybe I’m confused. – Thats your problem I live in Canada. In practice the questionable legality of libdvdcss in the S.U. sorry I mean U.S. is not a problem for home users no home user is ever going to be prosecuted for downloading it and using it to play their DVDs. On the other hand it could be a problem for a distributor that tries to distribute in th States.
Tax prep software – why run non-native under WINE? – Why not ? WINE is Not an Emulator and therefore does not have the speed hit of running software on an emulator, indeed some software runs faster on WINE than on Windows.
Cinelerra I can’t comment on. – Cinelerra I can’t comment on.
I don’t know who your users are, but they aren’t professionals in the universe I live in…you’re plain better off just running Windows
If I were a graphic arts professional I would rather use OSX than Windows. As a professional research chemist who is forced to use Windows at work. I can tell you I find no advantage for it over what I can do on my Linux box at home in fact I find Windows inferior in a number of ways.
2006-03-21 5:56 amormandj
“Exactly you said it. ” – Except you interpreted it in reverse. Oh well.
“Perhaps you meant Photoshop which runs under Wine ” – Like dogpoo the last time I saw it running. Half the plugins (really important…) didn’t function properly.
You can recite the meaning of the acronym, that doesn’t make it fast. “Indeed some software runs faster on WINE than on Windows” – Well, indeed, some runs slower. Most that I’ve attempted. The above mentioned being primary offenders.
“If I were a graphic arts professional I would rather use OSX than Windows. As a professional research chemist who is forced to use Windows at work. I can tell you I find no advantage for it over what I can do on my Linux box at home in fact I find Windows inferior in a number of ways.”
I would agree, OSX is a better choice. Linux is not. Not for GFX professionals. As a professional chemist, I would pick linux (actually, Solaris) as well. The right tool for the job. That was the point of my original post. Cheers.
2006-03-21 5:45 pmvoidlogic
libdvdcss is legal, read the wiki on it http://en.wikipedia.org/wiki/Libdvdcss
Well, I don’t see it as very objective, I think the author didn’t really go into enough detail, and was overly general in his analysis. That being said, I think it was an “ok” effort. I just don’t feel this guy really put much time/effort into it.
Edited 2006-03-21 02:17
“Ubuntu, Macintosh, Windows XP”
He compares two operating systems with a generic line of hardware products.
This was a pretty useless article that provided little or no insight into ANY of the products.
P.S. Yes, The GIMP is still waaaaaaay behind Photoshop
“I have found GIMP an adequate replacement for Photoshop and I just completed a 30 day testing period for the latest distribution of Photoshop.”
That is what you wrote, does that mean that you tested an upcoming version of Photoshop running on linux. If that is the case, are you allowed to reveal that because i cannot find any info about it on Adobes pages.
2006-03-21 4:40 pmricks1950
The GIMP/Photoshop debate is quite interesting … I’ve used both, and now use the GIMP. As an amateur photographer, it is more than adequate. I understand it lacks the pre-press CMYK colour matching, and the interface is different — not inferior, just different. It is suitable to task for me. Personally, I find not working like Photoshop kind of a bonus for using the GIMP! PS always kind of pissed me off.
I believe that many, many users would find the GIMP suitable as well.
If you absolutely have to match the colour output of a printed page to photograph; if you need one of the few other features only found in PS, then, by all means, open your bank account, and buy it — and live with the OS forced on you as a result.
I first put Linux on my PC in 2002, and kept XP in a dual boot configuration, thinking that I would need some of the programs to do what I do with a computer. Within 6 months, I was no longer using XP.
2006-03-21 5:12 pmFlatline
I agree that for most users GIMP would be more than sufficient, as long as the UI paradigm doesn’t bother them (it does take a bit of getting used to).
I think that most people could be pleasantly surprised by the free offerings available if they would give them a chance. Not all free software is good, of course, but there are an awful lot of high-quality offerings.
“Programing involves using Windows shell extensions”
What the hell is he on about? You can program windows forever and not “use” a shell extension.
I guess if you count going to explorer and right clicking a file and selecting send to… or using the tortoise SVN shell extensions or…
He lost his credibility there. If you don’t know what your talking about, don’t try to BS your way through it.