Before kids can read, any "conventional" gui (I don't care if it is KDE/Gnome/Windows/Mac) is both going to train the kids to ignore dialog boxes and/or drive parents to distraction with questions. That's why (in part) OLPC do the Sugar user interface: our target is kids who are in the process of learning to read.
It's also why the OLPC XO-1 is much, much more rugged than conventional laptops.
Batteries are a big profit center for companies. One of the things we worked hard on the OLPC to achieve is extended battery life.
You can trade somewhat lower capacity for longevity. Basically, if you are charging the battery, and take it to full charge, you are in fact damaging the battery slightly. So we don't fully charge the battery, so we can get many, many more cycles out of them (we use LiFE, batteries as well, which are much safer than LiIon.
So that extra 10% or so of "run-time" ensures you'll wear out the battery quite quickly, and you'll buy expensive batteries for the life of the laptop.
So you see marketing on how long your laptop will run, but not how long the battery will last. In our case, the kids may be literally days or weeks from anywhere you might ship replacement batteries to (presuming they aren't stolen on the way), even if they or their school could afford to replace them.
One of the parts of a low power machine such as ours is that our batteries can be much smaller and cheaper as well, if they do need replacement (or you want a spare).
Carla Gomez's trip report had prepared me intellectually for what I would find when I visited in mid-October, but nothing can really convey the emotional impact. Here are some of the notes I took talking with the teachers. This was about 4-5 months after the trial started, using our B2 systems which were much slower and much inferior software. Sometime, maybe I'll have a few minutes to blog about it.
Impact on students and teachers in Arahuay
I took as complete notes as I could talking with each of the teachers in turn (unfortunately, I forgot to get their names); translation slowed things enough that I believe the notes are pretty complete, though it may have also introduced errors. They echo Carla's excellent report, but are now months later:
Two children have come to the Arahuay school specifically because of the laptops who would not have previously attended.
The children are sharing much more: they take pictures and videos and share them.
The children are teachers too.
The teachers see much improved conduct. One child (who often arrives hungry) who has been sad and aggressive now loves to work on the OLPC. He is working more with other children and his behavior has improved.
One of the children has vision problems; is cross eyed and has one damaged eye (Carla will remember the child, I'm sure). Using the laptop has improved the child's ability to focus her eyes and work.
Another teacher noted that small children, ages 6-7, are learning much faster.
The web browser is the most popular/important activity, followed by the camera.
The activities they use most are the browser, paint, calculator, write.
The children use the internet to find information of interest.
One child, who is from Lima, has learned much in Arahuay and is very happy about the OLPC.
Another teacher said the children have changed: they have more concentration, mental ability.
The children's concept reception is much better than before.
Despite the use of US keyboards (all we had at the time), the children have had little problem adapting, and have figured out all they keys.
A third teacher said the internet is the most interesting.
The children are showing more abilities, are more creative, their behavior is better.
The children were selfish about the computers at first, but now share and discover with them, showing the teachers and other students what they have discovered.
Children who had previously been interested in power (bully?) have forgotten power and are sharing.
The children are showing better attention and organization.
Students are learning about the world, and now feel part of it. They are now interested in learning other languages, which they had not wanted to do before. Creating a web site on Arahuay has made them feel part of the world.
Impact on the teachers:
They have started to research topics on the internet and have practiced to use the computer.
The teachers have more ways to plan and improve the class.
Another teacher said the computer was wonderful for her. Information on the internet had improved both her and the children.
Their jobs are easier now.
One of the teachers asked for mind-mapper software, which they have used. We should install freemind on the servers and explore how feasible it is for packaging as an activity (it is Java based).
But the high point was the eight year old girl who came up to me shyly and gave me a kiss....
BTW, if anyone speaks Quechua or Aymara (or other languages), please help at: https://dev.laptop.org/translate/.
Yes, the anti-theft system will be disabled and you can install other things on it: both Fedora and Debian run on the system today: see the laptop.org wiki.
The field of astronomy has a long history (at least 30 years) of being able to set priorities for funding, sometimes making hard decisions between projects This self discipline helps it greatly in government funding when funding decisions have to be made between project and facilities; otherwise projects would get "peanut buttered" to death, with too little money spread over too many projects.
If the scientists won't give guidance to the paper pushers, then they do the politically expedient decisions; the hard decisions aren't possible to make in government unless there is political cover from the community itself.
This is clearly another in this series of reports, typically separated by 5-10 years.
That much of the silicon we're building *hasn't even taped out yet* (we're testing FPGA versions before they become ASIC's right now). Yet open source drivers for the hardware are already publically available (e.g. NAND driver, camera driver, SD driver).
- Jim Gettys
He chose the word "fat"; I'd use the word "bloated".
Too many applications are hemmorraging memory. e.g. Firefox, skype,
Too many applications are just plain huge; e.g. Open Office.
Too many applications do plain stupid things, like leak pixmaps in the X Window server.
Too many applications link against libraries they don't even use, causing gratuitous references to them, and slower startup times.
People have become downright sloppy. Our systems, even with.5 gig of RAM like my laptop, have to swap things out due to this sort of sloppyness. This should just not be necessary.
If you ever wondered why our intereactive response is unpredictable, just consider what happens if you have to start waiting on disk drives to page things out and in.
This is (mostly) fixable, if we just buckle down and realize we have a problem that needs to be fixed.
Jim Gettys
OLPC
Not to say there aren't other reasons already mentioned here, but the most common situation I've seen (and I got quite a few specs released out of DEC in the early '90's), is that the specifications often have a mixture of information considered proprietary (e.g. detailed logic design) and the programming information.
Seldom were there a nice set of documents, partitioned between programming and logic design.
So you have to track down the right documents, edit out the stuff that hardware folks are worried about, and get the right sign-offs. And you therefore have to deal with whatever random document preparation system was used by the engineers to do the documents.
This all turns out to be a fair amount of time, trouble and hassle.
So even if there aren't other paranoid concerns, it is harder than one might naively think.
***Please*** see the following web page for an explanation of the new facilities in the upcoming release! It is located at http://www.freedesktop.org/XOrg/X11R68ScreenShots
The original poster meant well, but did not include the explanatory text with the screen shots...
The SD consortium folks did not publish enough information for open source drivers. You had to sign various agreements to see the specs, etc.
However, this situation is now changing...
Our dear Redmond friends recently asked them for permission to ship SD drivers in source form; we (HP) said "sure, so long as open source drivers are possible". So this got them off the dime to open up SD implementations (at least the software side; they are keeping the mechanical and electrical specs locked up; they want to ensure interoperability of the hardware, and enforce it as part of the contracts you have to sign to get access to those specs).
So the programming specs are getting opened up; this should have taken place by now. This didn't happen in time for the Nx5000.
There is an SD driver developed independently on the iPAQ handheld for Linux from information that had already leaked out over the last several years; this needs further work for particular SD chip implementations. But it was problematical to distribute, at least by a member of the SD group.
At least it is now possible for do drivers, not possible in the past. I don't know how long it will take to get support done for a particular implementation; if you are interested, go for it!
- Jim
Jim would of course prefer to have the time to build serif italic faces; but the artificial obliqing (for most, but not all people) is preferable than having the faces indistinguisable or choosing a different family.
He did say if he somehow got the opportunity, he'd build them at the angle we use in fontconfig by default (I think it is 10degrees).
I believe you can tell fontconfig not to use the artificial obliquing if you want to.
- Jim
Having lived through one of Kevin's early penetrations (that of Digital Equipment Corporation) I can report that the damage was large.
The issue is not whether Kevin damanged the machines directly: as he says, he probably did not. The problem is that as a manufacturer, if the systems you build products on *might* have been compromised, you can no longer trust them, and get to rebuild them from scratch (get out distribution disks from before any possibility of compromise, reinstall from scratch, rebuild, and examine all source code that might have been edited). As a manufacturer, you owe it to your customers to be very careful about trojan horses, etc. You don't know all of what *might* have been done to the systems, and you certainly can't allow such things to end up in products shipped to end customers.
It is this cleanup of hundreds and/or thousands of systems (since you may not know exactly what has been compromised in an attack) that causes havoc and great damages to the victims.
- Jim
Slashdot Mirror
Before kids can read, any "conventional" gui (I don't care if it is KDE/Gnome/Windows/Mac) is both going to train the kids to ignore dialog boxes and/or drive parents to distraction with questions. That's why (in part) OLPC do the Sugar user interface: our target is kids who are in the process of learning to read.
It's also why the OLPC XO-1 is much, much more rugged than conventional laptops.
Batteries are a big profit center for companies. One of the things we worked hard on the OLPC to achieve is extended battery life.
You can trade somewhat lower capacity for longevity. Basically, if you are charging the battery, and take it to full charge, you are in fact damaging the battery slightly. So we don't fully charge the battery, so we can get many, many more cycles out of them (we use LiFE, batteries as well, which are much safer than LiIon.
So that extra 10% or so of "run-time" ensures you'll wear out the battery quite quickly, and you'll buy expensive batteries for the life of the laptop.
So you see marketing on how long your laptop will run, but not how long the battery will last.
In our case, the kids may be literally days or weeks from anywhere you might ship replacement batteries to (presuming they aren't stolen on the way), even if they or their school could afford to replace them.
One of the parts of a low power machine such as ours is that our batteries can be much smaller and cheaper as well, if they do need replacement (or you want a spare).
Impact on students and teachers in Arahuay
I took as complete notes as I could talking with each of the teachers in turn (unfortunately, I forgot to get their names); translation slowed things enough that I believe the notes are pretty complete, though it may have also introduced errors. They echo Carla's excellent report, but are now months later:
Two children have come to the Arahuay school specifically because of the laptops who would not have previously attended.
The children are sharing much more: they take pictures and videos and share them.
The children are teachers too.
The teachers see much improved conduct. One child (who often arrives hungry) who has been sad and aggressive now loves to work on the OLPC. He is working more with other children and his behavior has improved.
One of the children has vision problems; is cross eyed and has one damaged eye (Carla will remember the child, I'm sure). Using the laptop has improved the child's ability to focus her eyes and work.
Another teacher noted that small children, ages 6-7, are learning much faster. The web browser is the most popular/important activity, followed by the camera.
The activities they use most are the browser, paint, calculator, write.
The children use the internet to find information of interest.
One child, who is from Lima, has learned much in Arahuay and is very happy about the OLPC.
Another teacher said the children have changed: they have more concentration, mental ability.
The children's concept reception is much better than before. Despite the use of US keyboards (all we had at the time), the children have had little problem adapting, and have figured out all they keys.
A third teacher said the internet is the most interesting.
The children are showing more abilities, are more creative, their behavior is better.
The children were selfish about the computers at first, but now share and discover with them, showing the teachers and other students what they have discovered.
Children who had previously been interested in power (bully?) have forgotten power and are sharing.
The children are showing better attention and organization.
Students are learning about the world, and now feel part of it. They are now interested in learning other languages, which they had not wanted to do before. Creating a web site on Arahuay has made them feel part of the world. Impact on the teachers:
They have started to research topics on the internet and have practiced to use the computer.
The teachers have more ways to plan and improve the class.
Another teacher said the computer was wonderful for her. Information on the internet had improved both her and the children.
Their jobs are easier now.
One of the teachers asked for mind-mapper software, which they have used. We should install freemind on the servers and explore how feasible it is for packaging as an activity (it is Java based).
But the high point was the eight year old girl who came up to me shyly and gave me a kiss....
BTW, if anyone speaks Quechua or Aymara (or other languages), please help at: https://dev.laptop.org/translate/.
Please come help!
- Jim Gettys, OLPC
Yes, the anti-theft system will be disabled and you can install other things on it: both Fedora and Debian run on the system today: see the laptop.org wiki.
Incorrect.
OLPC continues to use LinuxBIOS: we use OFW as the *bootloader* rather than using Linux as Bootloader.
LinuxBIOS sets up the base machine, and then brances to a bootloader; so we are very much continuing to use LinuxBIOS.
The field of astronomy has a long history (at least 30 years) of being able
to set priorities for funding, sometimes making hard decisions between projects This self discipline helps it greatly in government funding when funding decisions have to be made between project and facilities; otherwise projects would get "peanut buttered" to death, with too little money spread over too many projects.
If the scientists won't give guidance to the paper pushers, then they do the politically expedient decisions; the hard decisions aren't possible to make in government unless there is political cover from the community itself.
This is clearly another in this series of reports, typically separated by 5-10 years.
Yes, as best we can. The keyboard is a rubber membrane keyboard (but much nicer than any you may have seen.
When closed, we intend it to be rained on or survive a dust storm.
No, not true.
We already had the alternate firmware project underway.
Theo has not helped the situation in the slightest.
RMS has been very civil in our extensive mail exchanges.
Theo de Raadt, on the other hand, has not been civil in the slightest.
That much of the silicon we're building *hasn't even taped out yet* (we're testing FPGA versions before they become ASIC's right now). Yet open source drivers for the hardware are already publically available (e.g. NAND driver, camera driver, SD driver).
- Jim Gettys
He chose the word "fat"; I'd use the word "bloated".
.5 gig of RAM like my laptop, have to swap things out due to this sort of sloppyness. This should just not be necessary.
Too many applications are hemmorraging memory. e.g. Firefox, skype,
Too many applications are just plain huge; e.g. Open Office.
Too many applications do plain stupid things, like leak pixmaps in the X Window server.
Too many applications link against libraries they don't even use, causing
gratuitous references to them, and slower startup times.
People have become downright sloppy. Our systems, even with
If you ever wondered why our intereactive response is unpredictable, just consider what happens if you have to start waiting on disk drives to page things out and in.
This is (mostly) fixable, if we just buckle down and realize we have a problem
that needs to be fixed.
Jim Gettys
OLPC
Not to say there aren't other reasons already mentioned here, but the most common situation I've seen (and I got quite a few specs released out of DEC in the early '90's), is that the specifications often have a mixture of information considered proprietary (e.g. detailed logic design) and the programming information.
Seldom were there a nice set of documents, partitioned between programming and logic design.
So you have to track down the right documents, edit out the stuff that hardware folks are worried about, and get the right sign-offs. And you therefore have to deal with whatever random document preparation system was used by the engineers to do the documents.
This all turns out to be a fair amount of time, trouble and hassle.
So even if there aren't other paranoid concerns, it is harder than one might naively think.
Both Croquet and LookingGlass are examples of innovation not found elsewhere, and are enabled by the new facilities.
The original poster meant well, but did not include the explanatory text with the screen shots...
I presume you mean the SD slot as media card...
The SD consortium folks did not publish enough information for open source drivers. You had to sign various agreements to see the specs, etc.
However, this situation is now changing...
Our dear Redmond friends recently asked them for permission to ship SD drivers in source form; we (HP) said "sure, so long as open source drivers are possible". So this got them off the dime to open up SD implementations (at least the software side; they are keeping the mechanical and electrical specs locked up; they want to ensure interoperability of the hardware, and enforce it as part of the contracts you have to sign to get access to those specs).
So the programming specs are getting opened up; this should have taken place by now. This didn't happen in time for the Nx5000.
There is an SD driver developed independently on the iPAQ handheld for Linux from information that had already leaked out over the last several years; this needs further work for particular SD chip implementations. But it was problematical to distribute, at least by a member of the SD group.
At least it is now possible for do drivers, not possible in the past. I don't know how long it will take to get support done for a particular implementation; if you are interested, go for it!
- Jim
This is a work in progress.
I wasn't expecting it to get slashdotted.
Roadmaps show you where you were, where you are, and maybe where you are going.
I plan to do more on where things are going...
And it would be good if other projects did roadmaps of their own projects.
The bits never leave the X server.
The current implementation is software only, and
runs at usable speed.
So I expect when we start using the alpha blending hardware, we'll run like a bandit...
The blending is being done entirely in the X server.
The speed with depend on whether the X server has hardware support for blending.
The prototype doesn't; it is, however, fast enough to be useable even doing it all in software with a Vesa server, so I think we're ok on performance.
Browser versions numbers have nothing
to do with the version number of the HTTP protocol.
The web today is using HTTP/1.1.
Please get a clue someplace.
- Jim
Plain text is easy to search.
And the HTTP/1.1 spec is available in
Postscript, with TOC and index.
- Jim
Thanks.
We tried hard to make the HTTP/1.1 spec
reasonable...
- Jim
There are words about gift horses here that
might apply...
In any case, Bitstream had never built
serif italic faces for Prima, Vera's progenitor.
- Jim
Jim would of course prefer to have the
time to build serif italic faces; but the
artificial obliqing (for most, but not all
people) is preferable than having the
faces indistinguisable or choosing a different
family.
He did say if he somehow got the opportunity,
he'd build them at the angle we use in fontconfig
by default (I think it is 10degrees).
I believe you can tell fontconfig not to use the
artificial obliquing if you want to.
- Jim
This was over 10 years ago (around 1988 or '89);
network security and tools was not what it is to day.
And you don't know if the attacker is malicious: you get to worry about "worst case" scenarios. You don't get to know what systems were compromised.
A white hat attacker is very different than Kevin's actions.
Net result is a very large cost to the organization attacked.
So I'm hard put to say he did us a service.
- Jim
Having lived through one of Kevin's early penetrations (that of Digital Equipment Corporation) I can report that the damage was large.
The issue is not whether Kevin damanged the machines directly: as he says, he probably did not. The problem is that as a manufacturer, if the systems you build products on *might* have been compromised, you can no longer trust them, and get to rebuild them from scratch (get out distribution disks from before any possibility of compromise, reinstall from scratch, rebuild, and examine all source code that might have been edited). As a manufacturer, you owe it to your customers to be very careful about trojan horses, etc. You don't know all of what *might* have been done to the systems, and you certainly can't allow such things to end up in products shipped to end customers.
It is this cleanup of hundreds and/or thousands of systems (since you may not know exactly what has been compromised in an attack) that causes havoc and great damages to the victims.
- Jim