Talk:School server

Status of this info

What is the status of the info in this page? User ramblings or OLPC page?

At first it seemed to fit the Category:Hardware ideas, but the use of XS & XSX names seems to hint or plays into some kind of officiality... although it doesn't give the impression of being so.

Given that there are OLPC plans for the servers, this page should clearly state the origin of its content: community or OLPC. --Xavi 09:34, 27 January 2007 (EST)

This page (the XS_Server_* pages) are currently ramblings (hence not referenced from a higher level), which over time have some decent chance of becoming official.

I'm new to these parts, can you explain why the origin of the information must be stated ? --wad

I was just wondering if my classification was 'correct'... :) I had heard 'rumors' about the server and its 'soon more info'...

Thus the 'origin' question: 'idea' or 'fact'? Personally, I take 'ideas' pages as that, ideas--without any concrete implications. While non-ideas pages have a bit more 'weight' on the subject. But that's my PoV... and of course, material is not consistently edited :) --Xavi 00:50, 28 January 2007 (EST)

All 'facts' begin as 'ideas'. You stumbled on a set of questions being posed, which I hope will blossom into 'facts'. The wiki is being used so that these questions can be discussed and edited by interested members of the community. --wad 1/28/07

The correct classification would also list it in Software ideas, as it is a system comprising both hardware and software.--Wad 00:03, 29 January 2007 (EST)

Any old OLPC laptop is a server

All of the OLPC laptops can act as servers. If one kid creates an activity bundle, then his laptop becomes a server. If one kid plugs in a USB thumb drive or USB CD Reader, then his laptop can serve content to others.

Of course. But there are resources needed which aren't well served by ad-hoc presence. If your laptop is acting as a server, you can't just put it too sleep to save power, or close it up to take home! Right now, we use a variety of methods for transferring content from machine to machine (rsync, ftp, smb). We can also do this transfer using an intermediary (rsync,ftp,smb it to a machine which serves it up via http/ftp). The advantage of using the intermediary is that others (even you) can easily access it later.

We ARE trying to build a system which is as distributed and scalable as possible (c.f. the choice of mDNS for service discovery). Perhaps the school server is just a laptop running a different set of software. At the end of the day, however, good engineering practice will probably suggest a slightly different combination of processor/memory/network than that optimum for a laptop, thus the discussion about school servers--Wad 00:49, 2 February 2007 (EST)

These pages seem to have been written by someone stuck in the US high-school IT environment mindset. Free your mind if you want to contribute useful stuff to the OLPC project! Who would have thought of a screen which shifts to low resolution monochrome to save energy? Who would have thought of a wifi module which keeps on running after the computer is shut down?

Please take your ad-hominem attacks elsewhere, they are not constructive... (And please identify yourself!) Constructive would be starting a separate page dedicated to explaining exactly how each of the services being proposed for the school server could be implemented in a fully distributed manner, and starting a real dialog on the pros and cons. Feel free to link to it from the School server page.--Wad 14:43, 1 February 2007 (EST)

BTW, the network interface of XO is strongly reminiscent of the Monsoon project at the MIT AI lab. Separate network processors are not a new idea.

School district networks

Indexing, whitelisting, blacklisting

A sensible index of suitable web pages for unlimited access by pupils in school could, for example, be built from the index of http://scholar.google.com/ (if available under license), the search index of the National Science Digital Library (if available under license) and further search engine indices (e.g. http://education.wikia.com/wiki/Search). A crawler could build a list of "neighboring pages" and permit websites into the index which were neither blacklisted nor filtered by automatic content filtering nor suspiciously "near" to blacklisted pages. Blacklisted pages could be pages without educational content but sufficiently interesting content so pupils could waste a lot of time on these pages. A small problem could be that some pages may contain games and edutainment adequate for younger pupils but inadequate for older pupils. A tolerant strategy would just allow these pages for all age groups. Use of a large web cache per school district or region would suggest itself for schools in developing countries. --Fasten 12:07, 21 February 2007 (EST)