Re: [f-cpu] Re: Welcome (late threads) (better late than never ;-D)

[Yann Guidon <whygee@f-cpu.org>]

hi !

cool, i realise that it is the first thread on this server :-)

Carlos De Urresti Hannoh wrote:
> From: "Yann Guidon" <>
> Sent: Friday, May 04, 2001 5:05 PM
yeah, that's "late" ;-)

> Subject: Re: [f-cpu] meeting in Paris this evening
> 
> > hi !
> >
> > The discussions about the server(s) are going on on both french
> > and english mailing lists (curiously, the german one is mute...).
> > i wish there was no langage barreer...
> How will the language barrer will be applied in the near future when we start to
> take priority changes on the project or know that we already moved to seul.org?
> (Will we apply the different languages lists again?)
the 'local' lists are necessary, they deal with local stuff and don't 'pollute'
the main list with local langages. You should have read the reactions
of people reading french messages on the english mailing list, they were
not very happy... :-/ Some people imagine that english is the only langage
on Earth and other langages are superfluous. At least, the german and french
mls help avoid discussing about this silly subject :-)

> In my point of view keeping those localized versions could help people that are not
> capable to express their ideas in a well written english form (i include myself ;>)
> to put in the project their thoughts and maybe someone could help them to translate,
> etc .. (of course this shouldn't THE rule, and everyone would be encoraged
> to write at the english list.)
that is more or less what is done.
fortunately, until now, people can read and write english,
enough to express their ideas.

it appears that when a "technical" idea is sent to a "local" ml,
the thread explodes and early decisions are often taken.
then, it takes a lot of time to explain what has happened, and
non-local people feel excluded from the discussion.
therefore, technical discussions or decisions are not good on the
local mls, they waste time and efforts.

> > Carlos De Urresti Hannoh wrote:
> > > > hello everybody,
> > >...
> > > > Phase 0 will be to find a location, a network, a legal and official
> > > > arrangement for the costs (electricity, climatisation...), putting
> > > This might help a little bit :) I work on the first IDC (Internet Data
> > > Center) in LatinAmerica. Giving a brief resume of the capabilities out of it are:
> > > (pretty fancy) Area of the Site: More than 7,000 km^2.
> > hmmm i think you mean : "7,000 m^2." :-)
> Sorry ;)
> wonder... what if i had write euro km ;)
dunno, it wouldn't sound realistic either ;-)

> > however, if we use Celaro hardware, we'll have to check if the exportation
> > is allowed to Mexico. The patent struggle with Quickturn makes selling Celaros
> > to US impossible. i'll ask the people at Meta Systems.
> oki. i'll start a local research.
well, it seems that we're not going to use that in the near future, so it's
not yet a problem.

> > > If you want me to start looking for prices just pass me down a line.
> > if you can ask the prices for a "decent" SUN (min 1G RAM and 20G HDD),
> > then just go ahead, it will be helpful in any case.
> 
> Regarding this information i'm posting what i got (yg mentioned that the sun
> option is almost out of our field, specialy when only a little group of
> people are willing to raise funds for this project. Maybe in the futurue.
> 
> As we know, USiii still hasn't been market to the servers field.. so the
> only way to go is a blade 1000.. vs ultra 80 is only a little bit more
> effort .. and time prices will be at our side.
> 
> I requested an aprox bill for blade 1000 , USiii 900, 18.2 GB, 1 GB module
> and i got this.. sighs.. at least those help us to give us a brief idea
> of the reality.. *which we might already known ;/ * all prices are US dollars ;
> besides the even wanted to charge for the
> English only OS media at 125! and 50 more for the other localized versions..
> When we can get it for free at sun.com ;) iso download and the localized
> versions also..
> anyway.. they told that they could rearrange the prices when I pick one of
> their packages.. but after denaying to buy them the media, monitor, and their
> crazy setup onsite fee(300).. i guess no progress could be done with them..
> or not a higher rate as getting all those free items from them..

obviously, getting a SUN from SUN is not going to work...
who would "own" the machine anyway ? we have to find a trick.

> blade 1000 (USiii 750, 512 MB, 18.2GB) - 10,995 (we can have 2 cpus)
> 512 MB upgrade (4*128MB) - 2,530 (i told them .. i'll expand.. sure.. with
> the 8 slots
> ocuppied..)
> video connector adapter hd-15 - 39
> subtotal: 13,564
> Standard Warranty (1 year): 396
> pseudototal: 13,960
> 
> ultra 80 (USii 450, 1 GB, 18.2GB) - 12,095
> Standard Warranty (1 year): 1,008
> pseudototal: 13,103
> 
> I wrote and maybe flooded with this..so trying to keep out some un-needed info
> should i post the usiii 900 prices.. or only keep them at my dreams? ;)

that's either that, or we try our hardest to make an old
server work again...

> > > > -----Message d'origine-----
> > > > De: Yann Guidon <>
> > > > More than two years ago, we tried to see if we could
> > > > put a PC on the net, connected to a FPGA board. It
> > > > appears impossible for practical reasons but i have learnt
> > > > a lot about emulators, they solve all the problems.
> > > >
> > > > Now, it's only the beginning of a new "project" because
> > > > we need to federate other Free HW projects, in order to
> > > For a starting place of the Free HW website I could offer a 256 k/bit
> > > line(not a lot.. but for starting web server is more than enough. At least I think
> > > so. But not enough hardware :/ unless I run the web server on a P200 with 72
> > > MB RAM.
> >
> > It is nice and useful, but now :
> >  - a root team must be carefully chosen
> >  - we need a separate domain name. we can setup a subdomain of
> > f-cpu.xxx (such as eda.f-cpu.org) but if this evolves as a trans-project
> > server (if we receive help from other Open Hardware and Free Hardware
> > projects) a separate and neutral DN is desired in the future.
> neutral organization will be the main problem.
sure, but up to now, we're alone. we have to get it done in a way or another...

> > > From: boucli27 at altavista dot net
> > > Subject: Re: [f-cpu] Which machine to use
> > > > I think we may have a premium class server for ~$15.000 --> $20.000 thus
> > > > we will have to launch a subscribtion
> > > > in order to get one of theses babies.
> > > >
> > > > I think I will be able to find $1.000 for this project.
> > > > Who's next ?
> > > May I? :)
> > yeah :-) [i can't, unfortunately]
> > who is in charge of collecting the funds ?
> For a starting point we could relay over paypal and someone with an account
> in a US bank since wire transfer's fee are so high and even higer to send
> them as a money order. And then redirect from there.
i didn't understand ...

> > I think that before we do everything, we have to :
> >  - make a press release on slashdot, usenet etc...
> I agree.. maybe the workshop in HAL will also make this explote. ;)
sure.

> >  - wait for new contribution proposals (so we can spread the workload)
> >  - ask for price lists and call for sponsors
> Looking for sponsors, explaining later on.
the explanation is the most difficult part... heck...

> From: "Yann Guidon"
> Subject: [f-cpu] about the EDA server(s) [quite longuish]
> > New project name, domain name + ml + CVS + etc. are necessary,
> > as well as a core team, admins, a central server, and legal existence.
> If we need to plan a root admin team, how would their members be selected?
> Recalling the search for admins at seul.org's mailinglist a question was exposed
> and some people raised their hands. i know that developing and choosing a root
> team members is a verify thing because of the responsabilities,.. any ideas?
let's first find volunteers for the seul.org ml :-)
then, let's setup the server.
before we achieve that, we'll find volunteers :-P

> > I don't think that it should remain F-CPU specific and other
> > projects would benefit from that. As a rule of thumb, several
> > projects listed in the OpenCollector database can apply and
> > participate.
> Thinking in the future and assuming that the neutral EDA localtion will be
> sucessful; the other projects could start thinking in contributing
> "semi-sponosoring" with hardware (relocating their own facilities to the EDA
> localtion and not fulling giving the ownership to the EDA project). when
> this stage is reached: how will the
> neutral position will be managed.. or we won't accept semi-contributions
> from other projects?
i don't know... but a contribution is a contribution, i don't understand
what a "semi-contribution" is...

> > It is, on one hand, a matter of usefulness :
> > if what we do is useful, i don't see why we shouldn't prevent
> > other projects from benefitting from our efforts, as long
> > as they are not "passive users" (they have to help).
> I fully aggre, beside isn't this the purpose of the fcpu as the GNU has done
> until now? :)
As you understand, today, this is not the only reason why we need other's contribution ;-)

> > B) Goal of the new organisation
> >
> > It must be defined when all the team members are present.
> ?
yeah, i mean : we can't decide before the team is created...
it will be adapted when people contribute etc...

> > It requires a clear, short and precise description of what
> > the organisation is, and is not.
> > The F-CPU project certainly suffers from the same problems
> > as other projects, so here's a first proposition :
> >
> > "
> > The goal of this project and organisation is to provide Free
> > Hardware developers with free online design tools, without
> > contraining him to buying anything or learn complex books.
> books? :)
yeah, you know, cut trees with expensive white pages...

> > The purpose of this project is to promote Free Hardware projects
> > and lower the entry effort for contributors. We want to
> > allow people to contribute without discrimination (either
> > monetary or expertise). The only constraint is that the
> > server users comply with usage restrictions and rules,
> > related to security and licencing (all the developped files
> > must be protected by the GPL).
> This will be only functional over the own fcpu servers..
i don't think so. first, it will help for the recognition of
the GPL as a Free Hardware Licence. Second, it will be a
criterium for the choice of the projects.

> > E) Contributors
> >
> > this project needs a root team and probably a surveillance commity,
> > besides the legal stuff. This means that it is a fully serious project
> > and responsible people are necessary. One will not be able to leave
> > the project on a whim.
> The flooding has shown the problem of a centralized or autoritative (one
> person) and the consecuences.. in this case relocating the list without
> the proper planning in advance.
> 
> Lets hope in the future we can balance the responsabilites and knowlodge to
> evade the: "project on a whim"!.
it's a tough everyday work...

> > I can't take the responsibility of this project : the F-CPU and
> > other things are already taking all my time. I could however help
> > on punctual cases, such as setting up a HW emulator (i have a rough
> > experience in this era).
> In this moment i can only help with the sys admin area, since i need
> to learn a lot .. at least reading the manual helps me to research
> for specific topics
Any help is welcome ;-)


> > F) Routes
> >
> > There are several and non-exclusive ways to achieve the project's goal.
> > These routes will probably be explored parallelly, it is a good way to
> > have at least something working if everything else fails.
> >
> Giving a brief resume.
> At my place we're trying to convince their univ. to bring back a supercomputing
> area which was desalocated in the late 80s, because none was developing
> or deploying it. The first phase is to give them an interesting method of the
> use of this area. At least on the Chemistry department a professor is willing
> to give us his support (one of the top ;>). I'm also working on a 'phases'
> document to give it to the univ. as the initial plan. And if this success
> our plan is to bring companies into the project as sponsors since many of the
> local companies need processing data out of their actual limits; expecially
> on the simulation area ;).
yeahhh :-)))

> As everything we need to start at the lowest level and currently i'm trying to
> build up a plan 9 lab but the cpu distribution gave me a bad look over it. I
> would like to see parallel clustering joined with a distribution clustering ;)
> Going back to reality.. we know of the huge process of programming over parallel
> but it'll a challange , not as fcpu, and hopefully some of the resources will be
> open for the cpu.. anyway we would have to write our own eda tools..(?)
> hmm..

i'd like to convince some sponsors in the EDA SW industry.
as you may know, there is no good GNU EDA SW and we can't
write our own (it's not the f-cpu project's goal).
i try to get in contact with Cadence PR.


> From: "Yann Guidon" <>
> Subject: Re: [f-cpu] about the EDA server(s) [quite longuish]
> > my short term todo list includes : cleanup of the Oekonux conference
> > files and preparation of files for HAL.
> Any 'public todo' list for HAL?

everything is to be done :-)

mantis.f-cpu.org should be up one day, it will help us
organising the project tasks more efficiently...


> From: "Yann Guidon" <>
> Subject: Re: [f-cpu] how to keep manual in other languages updated
> > however, the most up to date version is in LATEX
> > (you'll see, it's only a text file in fact, a bit like HTML)
> I do recomend you this application:
> ftp://www.ctan.org/pub/tex-archive/systems/win32/fptex/ .
> It will generate you a vdi file which can be read
> with the inclued viewer.  Also you can still go for the code ;)
i'll try to remember to link to it :-)

> From: "Yann Guidon" <>
> Sent: Saturday, May 19, 2001 4:46 PM
> Subject: [f-cpu] i should be sleeping instead of writing weird stuff, but...
> > It appears that the "first phase" of the EDA farm project could be
> > easily reached : we require a good athlon or P3 box with a large amount
> > of RAM, connected with a fast link to the net.
> 
> I can add another machine for the farm but my aprox time for getting it is
> like 3 months starting to count after HAL :/
> 
> But what will "farm" will actually mean aplied to EDA.. I'm not used to
> EDA.. We'll have each machine working on different tasks? or?

well, with 'farm', i meant more something like a "pool" of machines,
available for doing work. Because of the potentially high heterogeneity
of these machines and their geographically spread locations, it
seems that the 'farm' word (that is close to "clustering") was probably
bad.

> > A secure kernel with
> > all quotas enabled would run the compilations and simulations in a batch
> > queue. Of course, the licence for this server is not easy to get,
> > i need your help to convince Cadence and the others.
> What can we do to convince Cadence?
i don't know. I'm still waiting for their phone call...

> Von: Yann Guidon <>
> Betreff: [f-cpu] f-cpu.org/index.html updated
> Datum: Thu, 24 May 2001 03:21:53 +0200
> > hello,
> >
> > http://www.f-cpu.org
> > has been updated with some more acurate data
> > and a few new features. It's getting heavier,
> > a webmaster is still wanted...
> May i apply? :)
of course :-)
i hope that you have "some free time"...

> Orage.
> (Who do the work of @avanticorp.com at seul dot org) > Bad joke.. ;+>
:-)

WHYGEE who's loosing some hair on the head, wondering if it is natural...
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
*************************************************************
To unsubscribe, send an e-mail to majordomo@seul.org with
unsubscribe f-cpu       in the body. http://f-cpu.seul.org/