I'm getting very tired of trolls for religious wars about Microsoft XP versus Sugar-over-X-over-Linux.

Why don't we just port Multics (http://web.mit.edu/multics-history/source/Multics_Internet_Server/Multics_sources.html) to the machine and be done with it?

I make my living using and writing applications for XP. And I am a G1G1D1 donor who loves his XO and has written an activity for it (StarChart) that's been downloaded an average of once every 20 minutes for the last six months.

Oh yes, and once upon a time I wrote code for applications on Multics.

Uh oh, I'm getting sucked in by a troll...

"XP? Even grown ups have trouble liking Blue Screens of Death."

How much would you know about BSODs, mister "I haven't used their products for a very long time"? BSODs are very rare for me, and I've been using XP for 6 years. I get them once every two months or so at work, and never at home. By the way, have you heard of "kernel panic"? I expect not, since that never happens on Linux.

"an outdated system that remains unpatched at best"

What's this, Microsoft's custom XP XO OS is going to be the "unpatched" XP from 2001--at best? I'm surprised someone who doesn't use Windows could get such shocking insider knowledge.

Author wrote:

"55,000 Sugar/GNU/Linux XO machines are being shipped every month to kids all over the world."

Who let this clown write such imbecilic, lie-ridden article? I'm pretty sure his "proof" consists of taking some arbitrary number from one of Negroponte's lies/interviews and divide it by a convenient number of months...

Hmm, maybe I was to harsh up there. I sound like Irvin's opening act ;)

But I would leave the XP-bashing to the people who actually use XP, otherwise it sounds a tad biased. There are some things I hate about Windows, like its closed nature, but its stability isn't a problem (my impression: instability is usually caused by bad drivers not written by MS.) Its vulnerability to viruses is a problem, not because it is unpatched (which would be a problem on any OS) but because it lacks an effective security model and because it is so popular (making it a popular hacker target).

Obviously Office would be dumb for 5-year-olds, but I've always thought it would be nice if XO could run XP so that adults would have the option of using it. OLPC needs volume sales to lower prices, and selling an XP version sounds like an easy way to do that. Anyway, surely Sugar can be adapted to run on XP? Of course if you stay in Sugar all the time, it doesn't make much sense to spend the money to have XP running underneath, but that's up to the purchasers to decide, I figure.

About the only thing I agree with is the bit about Office being less than suitable for education. Even its value in the vocational perspective is a long shot, because the 10 year olds who use Office today are going to be using very different software 10 years from now.

As others have pointed out, a lot of the stuff about Windows is misinformation. I would also point out that a lot of the stuff about Sugar is misinformation. BUT NONE OF THAT REALLY MATTERS.

What matters is the value of the computer in education contexts. A child friendly user interface may improve its value, but it is not the end-all and be-all of educational value. What you really need is a diverse pool of educational software, software that addresses different subjects and different learning styles. Some of it has to be for math, some for languages. Some has to be skill and drill, while some has to be constructivist. Playing economic development favourites (e.g. emphasising math and science) is not much different from saying, "Office bad because it stresses vocational skills over other forms of learning. Ideological stances over teaching doesn't address the very real fact that different people learn in different ways, and different cultures have different ideas about how their child should be educated.

The final thing is that we know these things are being used in schools. If they are being used in schools, they have to make the teacher's life easier too. While the growing pool of education software for Sugar may achieve part of this goal, it is by no means the only thing that has to be done. Equally important are resources for teachers, so that teachers can direct the learning of children.

The most retarded article I have read all year.

So, XP has been around for seven years now and has, obviously, had a tremendous effect on education. It hasn't, just as Sugar in itself will not revolutionarise education.

However, if you would design a netbook for small children in poor countries, XP simply is bad in every possible way. From it's power hunger (slimmed down XP is sold nowhere) to it's business Desktop Metaphore, it has ZERO connection to the lives of primary school children.

There must be a better way for children to organize their favored activities (pun intended)? Sugar is one answer to that question.

And it is my deepest conviction that very little of value can be learned from MS software. Whatever you knew about 95/NT/ME/XP is useless in Vista.

If have yet to meet a long term Windows user who does NOT deeply distrusts or even fears her/his computer.

I have met *very* few long term Mac or *nix user who did not love and trust their computers. I am not talking about whether these feelings are justified, but they are there.

For children, an XP computer is the start of a very long abusive relationship between child and technology.

Winter

55,000 XOs per month are *shipping*? Do we have some independent verification of that? Also, since all those machine are going to governments and education agencies, what are they doing with them?

@allen:
"55,000 XOs per month are *shipping*? Do we have some independent verification of that?"

Does this mean you see this number as a political statement you should distrust. Does it really matter whether it is 40,000 or 80,000?

Given that Sugar is shipped with all XOs, the number of XOs shipped equals the number of Sugar installments shipped. Is there a problem with that?

If someone writes that X installments of Vista/OSX/Linpus ship monthly, why should I cry for an independent confirmation of X? Who cares?

Winter

Just for your information:

"Sugar is now in the hands of a half a million children and teachers worldwide, Bender says."

http://www.groklaw.net/article.php?story=20080920181151638

Now, if anyone would like to attack that number, please state evidence for Walter lying on this.

Winter

you know what i find funny? how so many people are against XP and yet one of the biggest subsections on olpcnews forums is how to install ubuntu onto the xo.

now i have little to no actual proof of this, but i'd wager a dollar that alot of the people who complain abut xp have either xfce or ubuntu or something which gives them a more standardized GUI on their xo. this, in my book, is hypocrisy. because for all intention purposes, ubuntu (with or without WINE)=windows. sure you can make all sorts of different arguments about how they aren't the same, but really, when it comes to it, the 3rd world kid who is going to use their XO in a manner similar to that of how the average kid in the first world uses their computer, the OSes GUIs are virtually identical. for the l33t kid who knows or want to know his stuff, there is nothing holding them back from installing linux. but for the average kid/user, it doesn't matter all that much

what sets sugar apart from everything else isn't that its linux or that its open source or blah blah blah, it's its creative gui. people complain about MS and how its originally created for businesses and what have you, but they turn on their XO in which they had installed ubuntu or what you, and use the same style gui that MS has.

but perhaps i'm wrong. perhaps the people who complain about MS really are just using sugar and hate the current/common which is set up on virtually every computer in the world. or maybe there are a lot of hypercrits who aren't really thinking about what they use/prefer when it comes to really doing stuff on the computer.

i am not here making an arguement for what they should/should not install for an OS on the XO. i'm just saying there's ideal, and then there is real life, and that i think when it comes to this argument about how terrible XP is and how great sugar is, that alot of people out there need to look at what they have installed on their XO before talking smack.

-my 2 cents

If have yet to meet a long term Windows user who does NOT deeply distrusts or even fears her/his computer.

Well you've just met one. I like XP, hardly ever encounter the blue screen of death and ...

I have met *very* few long term Mac or *nix user who did not love and trust their computers. I am not talking about whether these feelings are justified, but they are there.

.. loathe Linux with a vengeance. Oversold, overhyped and frequently software inadequacies are blamed on the user. People who invest the effort to get past this will love the OS. A self-selecting exercise.

Mark

When I first heard about the XO laptop I was very excited about the prfospects for the developing world to leapfrog directly to the front of the world information revolution. It initially solved a great deal of the issues developing nations face, and looked like a pure winner.

However, since every last shred of purity has been removed from the XO (operating sytem, independence, power supply and networking), and the 'reworking' of the aims of the XO, there really has been little to recommend it. Politicians and Organisers seem to have been bought off by the larger IT players, and those megacorporations that fear a new data economy based on ideas, not opporunity of birth.

You have sold the concept of the XO down the river, and allowed others to be bought off, too. The third and developing world's plight will I hope haunt the traitors the the ideal for the rest of their days. Being bought off has rendered those people directly responsible for continued poverty in those areas where the XO could have helped most.

Deaths will be on the hands of those crooked politicians and weasely traitors associated wit hthe XO's demise.

Damn you all.

@abe:
"i am not here making an arguement for what they should/should not install for an OS on the XO. "

We are, and we have the idea that an educational device for children who see their first computer might have different interface requirements than a administrative business worker. Just a question of ergonomics.

That means, that we distinguish between Graphical User Interface, eg, Sugar, xfce, or Windows, and OS, eg, Linux, Windows and OSX.

There are very good reasons to prefer Linux over Windows on a laptop for children. Security and long term stability are two of them.

This discussion is however about the best GUI for a child. I am convinced that a business desktop metaphor and Office applications are not the best way to support an educational environment on a very small screen for children that start off their use when they hardly can read. Sugar is developed for such children, and tends to be successful with them. As MS are not willing to integrate Sugar, or any other child directed GUI with their XP offerings, XP is considered unsuitable.

I am also convinced that Sugar is not the one and only, one-size-fits-all solution to computer interfaces. That is a very Microsoft look at computer software. Sugar is more along the Linux line of "everyone his due".

Winter

@Mark Tarver :
"Well you've just met one. I like XP, hardly ever encounter the blue screen of death and..."

That was not the question. The world's literature and every soap opera is packed with people who love what betrays and hurts them (eg, Stockholm syndrome).

The question was, do you *trust* your XP/Vista installation?

If yes, I clearly have not met you yet. Nice to hear from you. I am curious what brings you here? Please tell us, what is it that interests you in the OLPC? Interest in the developing world, education, technology?

@Mark Tarver:
".. loathe Linux with a vengeance."

Which again is not the question. I presume from your comment that they do not use Linux. Almost all the long term users I know trusted (and often loved) their systems.

Obviously, those people who hate it with a vengeance will not use Linux nor any other Unix variant long term.

There are people who hated Mother Theresa, Santa Claus, Gandhi, Mandela, and according to lore, every saviour and prophet who set foot on earth. And hated them with a vengeance. So I do not expect there will be anything it the universe that is not hated by someone. Not really a surprise then.

@Mark Tarver:
"A self-selecting exercise."

So why don't I meet these people who really trust their Microsoft software? The are so many who use it, but they all express utter distrust or fear of their computers. Where do these XP/Vista trusting people hide?

Winter

Trusting Windows or trusting Linux/Sugar really has nothing to do with it. Claiming that Sugar has a better user interface than Windows for children doesn't really have much to do with it either, since replacing Windows Explorer (the business like desktop that everyone criticises) is trivial.

What does matter is the computer's role in education. As an educator, I see Linux, and Sugar, as lacking a solid set of resources to fulfill that role. Resources include a flexible selection of software that children will like, and that can be shaped to the needs of the teacher and the curriculum. For the most part, reforming the tools has to be done at a high level, since very few teachers and Ministries of Education will have the resources (e.g. skill and time) to change the source code. On top of that, they need additional teacher's resources and maybe self-guided student resources which will allow people who aren't familiar with the technology use it effectively.

To a large degree, you can do that with Windows. Yes, they will have to pay additional money in order to use commercial software. But at least the option exists, and Ministries of Education that can license hundreds of thousands of copies at one time can probably cut a sweet deal. (Maybe more-so if they offer services such as internationalization of the software.) On top of that, there is nothing that excludes open source software from running on top of Windows. Squeak is a good example. Many Python activities will also work, provided you remove the dependencies upon Sugar specific API. Other stuff, such as electronic resources, don't really have a dependency upon open source or commercial products. Books can still be distributed as plain text, PDF files, or any other format that can be used in either environment. And yes, one of the roles of the XO is as an eBook reader.

The important thing here is: what will contribute to a child's education? Everything else is just a distraction that diverts everyone from the goal of a better future for those children.

I hate Sugar. Maybe it will evolve into gold but I hate it at present. Still, I'll take any open solution that is free and can evolve at the demands of the people over anything beholded to some private company.

Sugar is a GUI for children, not adults.

Its totally left field of every other 'windowed' GUI in existence.
Its geared towards collaboration not isolation as all other GUI/OS are.

For a tool to learn with, the XO/Sugar combination has always been right on the money. Its like giving a child Lego or Mecchano and watching the amazing creations they build.

You dont get that effect from Windows. Its just not designed that way.

(Banging head against desk.)

You can scream that Sugar is for children, not for adults, as much as you want -- but you are still missing the point.

People want more from their computer than a user interface. They want useful software. Sugar has software, but most of that software are gimmicks. By gimmicks, I mean you use it once or twice then never use it again. Distance is a rather extreme example of this. It's like: oh cool, if I have two very expensive computers I get a not-so accurate measuring tape that will only work in quiet-ish environments. You talk about the principles and say how it relates to lightning, then you're done with it. Measure* is pretty much the same, only you may be able to drag a few lessons out of that one because watching audio wave-forms and relating it to how we make sounds or how we perceive sound. On the flip side you have good activities like Write, because Write allows kids to both learn and create. The simple act of creating things extends the value of the software because the child is no longer a passive consumer.

Now Sugar has nothing to do with this dicotomy of the user as a consumer** and the user as a creator. Sugar has nothing to do with collaboration, besides creating a consistent user interface for it. Windows can do that stuff too, and it already has creative and collaborative software available for it. This software goes well beyond what's available for Sugar. It goes so far beyond the stuff that's available for Sugar, from both the perspective of creativity and education value, that it is a bit irrational to suggest that kids should use Sugar because the GUI is better for children.

Besides, do we really know that the GUI is better for children? I haven't heard of any studies backing this claim, so the people who are making the claim are probably doing so because: (a) someone else said so, (b) it feels correct because the UI is much more visual and much less cluttered, or (c) they have actually looked at usability studies that studied specific ideas rather than an integrated system. Well, claims and feelings don't count for much -- while studies of components rather than integrated systems can be misleading. In short, the GUI claim doesn't hold water until we see real usability studies.

* I shouldn't be too hard on Measure since I love that Activity. But the simple fact of the matter is that most people won't get it.

** Granted, sometimes non-creative skill and drill type software is desired. I don't much desire it, but my desires are beside the point as long as the other methods are still effective.

@Jordan:
"Sugar has nothing to do with collaboration, besides creating a consistent user interface for it. Windows can do that stuff too, and it already has creative and collaborative software available for it."

But not with the Mesh collaboration. That is what is meant. And XP collaboration is centralized. I do not think every school should buy a Sharepoint server.

@Jordan:
"Besides, do we really know that the GUI is better for children?"

Actually, we do. Child educational ergonomics is not some black art in the same field as necromancy.

Sugar was designed by people who actually are experienced with working with and studying of children. However, XP's GUI was designed (actually, ripped off) by people experienced with working with and studying of Office Clerks.

So, as a first guess, which do you think will fit primary school children more?

I think it is rather daft to insist that an office desktop GUI must be better for school children than something actually designed for them. You don't give them adult books (eg, about accounting) in school, so why give them the control panel?

Winter

Excellent comment, Jordan.

@Winter: But not with the Mesh collaboration. That is what is meant. And XP collaboration is centralized. I do not think every school should buy a Sharepoint server.

The mesh is a fairly low level protocol (either hardware, or very low level software) that creates a network of wireless devices. It has nothing to do with collaboration, short of providing the underlying infrastructure. SharePoint is a high level protocol, really a server, that allows for collaboration. In this sense, SharePoint is more analagous to the collaboration facilities provided by Sugar -- albeit, it takes a rather different approach. Since SharePoint and Sugar use high level protocols over TCP/IP for collaboration, there is nothing preventing decentralized collaboration tools for Windows (or centralized ones for Sugar/Linux). I would imagine a central reason for centralization through SharePoint is to improve discoverability (it is really hard to find other computers using TCP/IP unless you are on the same subnet, except with centralized servers). Centralized servers also remove the need for clients/peers to be always-on.

@Winter: Sugar was designed by people who actually are experienced with working with and studying of children. However, XP's GUI was designed (actually, ripped off) by people experienced with working with and studying of Office Clerks.

On the latter, not really. Concepts like folders existed long before the GUI. They were simply called directories. Folders don't really behave real file folders anyway. Back in the olden days we had filing cabinet drawers and file folders. A 2 layer deep hierarchy (with, perhaps, dividers adding a 3rd layer). Not only do the metaphores differ (folders are more abstract) but the flexibility of the metaphores differs (computer folders are more flexible). Anyway, there are all sorts of inconsistencies between how those office clerks organized the office and how stuff is organized in a computer to question your thesis and suggest that user interfaces are structured the way they are because they serve a broad range of needs.

Of course, that is mostly beside the point. First of all, we aren't talking about a broad range of needs. So you are right, we can probably use something else here. It is also beside the point because Windows applications (including a Windows Explorer replacement) can pretty much define the user interface to be what the developers want it to be. If that means Sugar-like, they can. If that means some sort of hybird between Sugar and the "desktop metaphore", it can. And it usually is. The activities are what's important here, not the program launcher and file organizer.

@Winter: So, as a first guess, which do you think will fit primary school children more?

Intuitively, I would say Sugar. However, just because something feels correct doesn't mean that is is correct. It is also worth noting that a good interface gets a job done, but a great interface is just icing on the cake. Windows is good, Sugar is the icing (assuming that it is proven). Since Windows is good, all of this quibbling over Sugar is nothing more than a distraction.

@Winter: You don't give them adult books (eg, about accounting) in school, so why give them the control panel?

Well, I would give a child an appropriate adult book. And while I may give a child a book with more child friendly content (suitable subject matter, suitable use of language, maybe a few more pictures), I wouldn't redesign the physical nature of the book itself (the use of bound pages, sentences and paragraphs, etc.) just because someone said that pages and sentences were not suitable for children. Children can learn. And sometimes children want to learn about those innocent adult things that we take for granted (like reading real books). Other times, we want to teach children real skills by using adult ideas, like writing proper sentences and grouping ideas in paragraphs.

In the case of the XO, the activity is the suitable subject matter, suitable use of language, pictures, etc.. It should be suitable for kids. The activities are the important things. The user interface is analagous to bound pages, use of sentences and paragraphs, etc.. It is not really important to change that stuff in the former case. It may be detrimental to the child's learning in the latter case.

@Jordan:
"In the case of the XO, the activity is the suitable subject matter, suitable use of language, pictures, etc.. It should be suitable for kids. The activities are the important things. The user interface is analagous to bound pages, use of sentences and paragraphs, etc.. It is not really important to change that stuff in the former case. It may be detrimental to the child's learning in the latter case."

It looks as if we agree almost completely.

Note that I used "XP" to stand for both the OS and the GUI. The point is not XP, the OS (which has its own drawbacks), but XP the GUI.

My biggest beef with XP the GUI is that MS is NOT going to change it. They could add Sugar, or a look-alike, or something better on top of XP the OS and that would bring it in the same league as Sugar + Linux (in this respect at least). There are good reasons to be very suspicious about a "Sugar shell" on top of XP the GUI. And I am very suspicious.

But to MS, XP is end-of-life. The do not WANT to invest any more money in XP.

The user interface is important in as far as it hides, and manages, a lot of the complexities of the computer. Although both children and adults read/learn from books, the layout and structure of the texts+figures is completely different. As is the language used.

Given that even most adults get lost in the menu structures of XP or MS Office, I do not see why children should be productive in it. And buying activities for XP should be no different than buying activities for Sugar: If you want to pay, someone will write it. As existing XP educational programs are hardly targeting, eg, Aymara children in Bolivia, I do not see such a large disadvantage in using Linux.

But on the whole, I simply do not believe that MS will change the interface of XP to suit children and education. They never did before, when XP was "hot", so I do not see they will do it now, when support for XP is about to die.

Winter

@Winter: "It looks as if we agree almost completely."

I think that it is best to say that we agree on the philosophy, but not necessarily the implementation. :)

@Winter: "My biggest beef with XP the GUI is that MS is NOT going to change it."

There are two comments on that. The first is that Microsoft has to develop a general purpose user interface that suits the needs of a wide variety of people. You don't need to do that with the XO, because the XO is (in theory) an education machine.

The second comment is that there is no more of a need to follow Microsoft in terms of their dictate for the GUI. People can and do modify it, and Microsoft doesn't do much to stop that. At a trivial level, companies like StarDock offer products that provide new skins and (at a shallow level) changes to how you interact with the system. Digging up long past history, there were a lot of products that replaced the Program Manager in Windows 3.1. You don't see as many of those with Windows 95 and later simply (I think) because most people are satisfied with the interface of Windows 95 and later.

But I think that's a moot point. If the stories about governments wanting XP for vocational purposes, rather than to access education software, is true then they won't want a UI that is vastly different from that provided by XP. I think that's the worst possible way to look at things, but I'm not the one spending the money here.

In some respects, I think that writing new software to target the needs of Bolivian children doesn't have much of a point either. At least if you are implementing something like this at the level of national governments. It is nice to think of developing culturally sensitive and child appropriate activities, but I don't think that the governments in the respective countries really want this. While I cannot speak for Bolivia, there seem to be pushes to standardize education standards on an international scale. This means that most people gain the same skills at the same time and perhaps even by the same means. It is a sad reality, but governments tend to see things in terms of economic competitiveness rather than social needs.

@Winter:
> Sugar was designed by people who actually are experienced with working with and studying of children.

By saying "people", who do you have in your mind?

@Yoshiki:
"By saying "people", who do you have in your mind?"

Now you got me.

I remembered reading that Alan Kay and some of his co-workers were involved in the Sugar design.

After trying to dig up the links again, I see Alan Kay is involved on the programming side of the OLPC, not Sugar. An I cannot find a history of educational research for those mentioned developing Sugar. But then, educational experts with good developing or design skills (or vice versa) are rare so they won't easily show up on the development team list.

So, indeed, the sugar team themselves might not have the educational expertise and experience, and I have no certainty whether or not they did consult experts. As I wrote, I think I read somewhere they did, but I know my memory fails me at times.

So, we will have to ask Walter Bender about their input for Sugar.

Winter

> I remembered reading that Alan Kay and some of his co-workers were involved in the Sugar design.

I'd like to read that, too.

To the design, I can tell you that Alan and his co-workers weren't *involved* in any real way. (I'm one of them, if you aren't aware of it.) In the summer of 2005 and other times, we occasionally went to Cambridge and had meetings with various RedHat and OLPC folks. But I don't think we contributed any meaningful influence on design. Their idea back then was more or less making a cheap Internet Appliance; you can tell that by looking the early prototypes here:

http://wiki.laptop.org/go/Sugar_demo_2

(Something like this was shipped with early prototype hardware, and I couldn't believe what I saw there.)

I believe Walter wasn't involved at that point (correct me if I'm wrong). The donut view and other ideas came after that, but I always feel that the initial ideas had set the course.

Our group tend to think that computer is a new kind of dynamic media, but the idea of "making/doing with computer" was totally missing at the beginning, and it is still weak in today's Sugar.

@Yoshiki:
"To the design, I can tell you that Alan and his co-workers weren't *involved* in any real way. (I'm one of them, if you aren't aware of it.) In the summer of 2005 and other times, we occasionally went to Cambridge and had meetings with various RedHat and OLPC folks."

"Having been there" always trumps journalistic reports.

Having not been there at all at any time, I just have to get by reading whatever information is made available.

Thanks for the info.

Winter