Linux, to be (Like Microsoft) or Not to be?

David writes "Stephen Shipman delivers a very articulate and concise view of how Linux fits in server and end user environments. He expresses his view in response to Nicolas Petreley's 'rant' in Linux Journal. He points out the subtle implications of efficiency versus consistency." From the article: "[...] efficiency (as measured by keystrokes) isn't the only metric for ease of use. Consistency must also be taken into account. Microsoft has made a lot of hay (and green) by flogging consistency".

Comments

Petreley makes good points

[ackthpt]

Microsoft doesn't get it. There are things in Windows XP which are still as idiotic as ever. This isn't evidence of a superiour product, but the result of understanding. The Registry is once again a completely backwards way of contending with things, and worse, you sometimes have to get into the Registry to change things which should be straight-forward options in personalising your computer.

Then there's the Single User aspect, all over again. No matter how they pass XP off as a multi-user environment, it carriest considerable baggage of being single user - case in point: the pop-up key-stealer, when apps suddenly thrust themselves forward and steal a keystroke for the [ignore] [retry] [cancel] [OK] whatever prompt and vanish if it meets the input expectation.

What I repeatedly hear from Mac enthusiasts is how quickly a new user can sit down and get right to business, without thinking half as hard where things are or how settings work. Microsoft made a big deal out of bringing a tonne of people on board to advise them and examine their user interfaces, but I grow increasingly skeptical that these were actually people flown to a nice resort, given fine amenities and still shown what Microsoft thought they should see, rather than simply gaining some real inside, i.e. "so what's the thing you most dislike about Windows/Office/Etc.?" Rather like a homeless guy will be your best friend if you give him a few bucks.

Consistency must also be taken into account. Microsoft has made a lot of hay (and green) by flogging consistency".

They also have become extremely overconfident because success came too easily. Note many of their recent failures. And may I be among the first of many to recognise Origami as an utter flop. Looks neat, but it's a niche player, same as Tablet Computers. It's too big and too small at the same time. Once again a complete misunderstanding of the market.

Linux should strive to be efficient and easy to use, not mugging one of the most inexplicably frustrating environments ever.

Re:Petreley makes good points

[ackthpt]

Just one comment, how can you say Origami is a total flop - AFAIK there's no units out there to buy yet at all. Just some hype on engadget etc, then some disappointment when it wasn't what they thought it was...then some interest when they saw the new interface.

It was years ago that I bought into every shiny new wizzy tech that came along. It took years to wear away my blind otimism that new==better. After spending a good amount of my own money and many long hours fighting with things to make them do what I needed experience etched it's way into my assessment of new, wizzy tech. I don't mean to come across as cocky or smug, but I think I've got to the point where I can take a look at something and determine if it's going to be useful and easy to use, or another exasperating time fighting with it to do what I need, not what the designers thought i should have.

One of the reasons I like being a programmer is writing my own tools. There are tools which will kinda-sorta do the things I need, but often more or less and not quite what I had in mind.

I look at Origami and see effectively a big Palm Pilot or smaller version of a Tablet Computer. It will no doubt be popular with anyone a laptop, tablet or Palm/PocketPC doesn't quite work for. On the last few flights I've been on and the last few conferences I've attended I have seen zero Tablets and few, if any, PDA size tools. Everyone hauls around a laptop. I think that's a pretty clear indicator of what the general population is drawn to. Origami is simply Microsofts misguided way of telling people, We know what you really need, despite many tools like this over the years which have vanished. Maybe UPS and FedEx will adopt them, but what they use looks like it could be run over by a truck and still function.

I don't think it's healthy to pattern user functionality on the designs of a company which is trying to expand into everyone elses market, instead of cleaning up their own back yard.

Re:Petreley makes good points

[johneee]

Yeah, this happens to me on Windows too.

I wait for something to complete downloading in IE and go to something else, and when IE's done, I get the thing on the taskbar flashing orange. If several IE windows are open, the rollup button flashes, and when I click on that, the one that wants my attention is flashing on the pop-up. No focus stealing involved.

Re:Petreley makes good points

[Hosiah]

that makes it so difficult for Linux to gain acceptance.

I don't give a damn if anybody else but me uses Linux, ever.

It doesn't have to "work for you", it has to "work for THEM",

No, it doesn't have to do anything. If it works for me, it's good for me, and if it doesn't work for them, they can go find something that does.

and if it works for you too, then that's even better.

Why would we be having a problem if there was a one-size-fits-all solution? I know for a fact that 90% of users out there would die screaming if they had to deal with my machine the way I have it set up - but it's how I like it. See above argument. Since when did Linux stop being about choice and start being about grouping everybody into yet another bunch of Borgs? I made my choice. It's Linux. Others made their choice. It's Windows. So be it. I don't crack into Microsoft's code base and rebuild the MS system to be exactly like Linux. I don't see where a bunch of Windows zeolots have any more right to piss in my Linux so it tastes better to them and ruins it for me.

FOSS Means Business, in Belfast, March 16th

[H4x0r Jim Duggan]

There's a conference this Thursday, March 16th in Belfast called FOSS Means Business where Stallman and Perens are both doing business-orientated lectures, plus presentations by Google, Open Source Academy, and Oracle.

People trying to encourage IT decision makers to transition to free software have to learn to explain it. Bruce Perens is good at this, but as well as telling people about the value of free software, we have to tell them how to hang on to it - how to not let it slip through their fingers. That's Stallman's angle, as can be read in this transcript of his lecture on GPLv3.

Microsoft isn't top because of their software quality, and free software won't displace them purely based on quality either. We'll win for other reasons.

They may have "flogged" consistency, but...

[Svartalf]

Considering that there's really not been any real consistency throughout MS' product offerings or
anything else about Windows' operating environment:

- Printing that doesn't work the same from Windows 95/98/Me to NT/2000/XP because of different
driver rules at the GDI layer.

- API's that change from one ruleset to the next without warning (the move from 16-bit to 32-bit
generated at least several API calls that produced nasty results because they used zero as the
default but in the 32-bit version they used a string for that parameter and they didn't account
for this in the API...)

- Consumer WinCE devices being allowed out the door with missing functionalities (i.e. The Uniden
UniPro 100 PDA was missing the Finder and a few other things- for no good reasons other than they
were short on firmware memory because of the added recording functionalities- and instead of
increasing the BOM costs slightly for more ROM capacity, they opted to omit some of the functionalities
that make it consistent with the other WinCE devices.)

- Apps don't have any consistent install/uninstall interface. (While Linux IS better in this regard,
it's got many of the same problems...).

- Apps often install their own DLLs to prevent being hosed by other apps and Microsoft when they do
updates.

There's tons more. "Windows" only seems consistent because the end-user community sees something that
"works like Windows" and is therefore familiar- since it's familiar, they whitewash over all the
issues about consistency and it "being easier to use". Issues that plague them day in, day out.

Microsoft may talk the talk, but when the rubber meets the pavement, they're not walking the walk- not even close.

Foolish Consistancy

[Hillgiant]

In order to assert that Microsoft has made a lot of green off of consistancy over efficiency, Microsoft's programs would have to be a lot more consistant. I hate hate hate that ctrl-tab does NOTHING in Word. UI options are hidden all over the damned place and only some of the settings are stored in the user directory (making portability a nightmare).

Yes, it's all true, and it's BIG

[H4x0r Jim Duggan]

The above should be flagged "sarcastic" for those who happen to lack such a barometer internally

No, it's true actually. A lot of businesses in Northern Ireland were poking at free software but no one wants to be first, so we're organising a big free software conference aimed at businesses. Stallman's name is a big draw. He knows it's a business audience and he'll adapt to that. He'll be including a substantial section about GPLv3, which has gotten a surprising level of interest from public administration bodies.

Interest has been huge and there were many requests for speaking slots that had to be turned down. I guess there will be a FOSS Means Business 2007 too, but one at a time. On Thursday we expect at least 300. The venue can hold up to 1150.

I think events like these, and the networking that happens at them, is more important than increasing efficiency of the software. We'll see.

Re:Microsoft

[The MESMERIC]

Linux is not hard for beginners to use.

I've installed Linux Desktop on laptops belonging to people that doesn't even know what an operating system is - and they got on well.

Now if you wanted to say Linux is hard to administer.

They yes, you are completely right.

If we get pre-built desktop system with Linux installed with all the compatible peripherals .. then I am of the feeling Linux is actually easier and less intrusive.

When you get a machine optimized for Windows, non-compliant BIOS, Linux-unfriendly video-card, broadcom wireless chipset, some Lexmark printer, some Canon Scanner .. some USB broadband modem, then you may find the prospect of installing Linux and trying to get everything as functional - as enjoyable as cutting your own eye-lids.

Not being a propagandist at all -- but one thing I find curious about non-tech users after they been exposed to Linux (for a while) .. is that they actually miss it once it's gone.

The Linux system obviously would have to be set in a very friendly way for that to happen - which is what I do to other people's dying PC when I am bored :)

Linux useability?

[dedeman]

Ok, I'm neither an expert using MS (take your pick) or Linux (take your pick). In an attempt to diversify my understanding of Linux, I started using RedHat 7.2, many years ago.

It was a slow, long, widing road, but I've learned, using a certain amount of perseverance.

It is the perseverance that the "average" user is lacking. Tell me how many of the following terms/words the shopper going to Best Buy or Circuit City are willing to learn: Source, Binary, Compile, RPM, apt-get, x86, X11, /etc/blah/blah, port(ed), API, drivers.

There are more, but I can't think of anything right now that would add to user/consumer confusion when all people want to know are things like "Can I use the internet with this", "I need some word processing", or the more experienced user that know that a hard drive size is measured in bytes, and the processor speed in herz.

Microsoft makes many things automated. Want OS updates? Go to windowsupdate.com, or click on the "Windows Update" icon. Want driver updates? Go to manufacturer, get drivers for 2000/xp OR 98/ME. No pointing to mirrors, no compilation, no source, no RPM, no Yum, just "Do It Now!", wait for the icon to appear, double click, make a sandwich, reboot.

That's what Linux is lacking. Does anyone realize this?

As Usual, Everybody Is Wrong

[Master of Transhuman]

I mean, both sides are as usual exaggerating the value of their positions.

What matters is functionality and usability. If Linux can match Windows in functionality, and if Linux is easily usable, it doesn't matter whether the technigues used are the same as Windows.

It only matters from the viewpoint of those people who wish to lure Windows users into using Linux. While it is true that most people, as one of my instructors likes to say, "use computers because they have to, not because they want to", this doesn't need to have any significant effect on Linux adoption, provided that the functionality and usability are there. Re-training is not that big a hassle IF properly done.

Most corporations are not going to switch to Linux just for improvements in usability or even functionality. They are going to switch for other reasons: cost, security, flexibility, lack of vendor lock-in. They will only switch for functionality if that functionality is mission-critical. Once the decision is made, people will either be re-trained or required to learn the new systems themselves.

Comparing vi and Microsoft Word on keystrokes is abysmally stupid. Vi is an overly complicated mess of un-usability. The learning curve is so ridiculous that nobody but a geek would even try to use it. The same applies to Emacs. Neither of them is intended to be a word processor, which is by definition designed for end users, not geeks. Even if Word needs more keystrokes than vi to do a particular task, this says nothing about why those keystrokes were chosen. While I wouldn't doubt that Microsoft designers are less capable of designing efficient keystrokes than Linux designers, just comparing the keystrokes doesn't tell you why it was designed that way. There may have been good reasons for using those particular keystrokes. My point is that comparing two totally difference systems - even if the function being compared is identical - based on keystrokes is utterly irrelevant to the usability issue, and by definition irrelevant to the functionality issue.

There was recently an article elsewhere about how GIMP wasn't as good as PhotoShop. As usual, everyone said it didn't need to be as the GIMP developers didn't care about that, and further, that no one had the right to ask that GIMP be equal in usability to PhotoShop as that was abrogating the rights of the GIMP developers to go their own way.

This is incorrect reasoning. The issue is whether GIMP is intended to be the best graphics program in terms of functionality and usability. The second - and different - issue is whether it can be recommended to Windows users as a replacement for PhotoShop in order to lure Windows users to Linux. The two questions are entirely different. If the GIMP has functionality and usability problems - and it does either when COMPARED to PhotoShop or in some cases on its own merits - then it should be changed to solve those problems . Whether the GUI is changed to look like PhotoShop or not is not relevant EXCEPT to those people on Windows who don't want to learn a new GUI. THAT is not the GIMP developers problem, clearly. But if the GIMP developers do not INTEND to develop GIMP to the same level of usability and functionality, they should say so, and people should then stop recommending the GIMP as a replacement for PhotoShop.

It does OSS no good to recommend OSS products that do not adequately replace their Windows counterparts. It's okay to recommend OSS products that are less functional for those people who do not NEED that extra functionality. It is not okay to recommend OSS products for those people who DO need that extra functionality. Saying that GIMP is a replacement for PhotoShop without specifying the limits on functionality and usability is not helping OSS because when the faults are experienced, the new user will feel cheated. Any recommendation of OSS software to users of other software should acknowledge any significant differences in usability or functionality. That is, if the product doesn't do a certain thing, say s