Server room clearout

[Simon]

A while back I mentioned some old x86 servers, HP and Sun, that were
due to be scrapped and there seemed to be enough interest that it is
worth my effort lugging them across town

I will be dropping off a HP DL360 G5 this evening.
It almost certainly has 2 CPU chips, but whether they are dual or quad
core I won't know until it get's powered up. Memory is probably 4G
minimum and maybe upto 16G but most likely 4-6G.
The disks, which have been completely wiped, are 2 * 72G SAS.

If the space can't use it then I am sure from the level of interest
shown a couple of months ago that it will be claimed and gone within a
matter of days...


Q. Should I put a 'do not hack' sticker on it for the space to use or


I have 4 more of these which should be available within a few weeks -
2*3.0GHz dual core CPU, 6G RAM
Q. who wants these when they are available?


Simon

[Paddy Duncan]

I'm up for 1..
Paddy

[Martin Klang]

On 1 Jun 2012, at 19:10, Simon wrote:

> I have 4 more of these which should be available within a few weeks -
> 2*3.0GHz dual core CPU, 6G RAM
> Q. who wants these when they are available?

that's my hand in the air -
we could certainly use that sort of grunt upstairs.

/m

[Billy]

I'd like one, if they're still going.

[Josh Miller]

im totally up for one
josh miller

[Josh Miller]

Could I please have one:)
Josh Miller

On Jun 1, 7:58 pm, Martin Klang <m...@pingdynasty.com> wrote:

[Nin Lil'izi]

Ai,
I'm still after a server for personal home tinkerings

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[Aden]

i'd like one please!

[Anthony Evans]

I'm definitely interested.

Anthony. 

[Adrian Godwin]

y'all know there were just 4, right ?

[Martin Klang]

runner-ups can have dual xeon DL360 G4's, if they like.

/m

[tgreer]

I'll have a DL360 G5 please!

[Anthony Evans]

So will I if there's one going.

Anthony.

[Matthew & Anna Israelsohn]

If there is a HP DL360 G5 available, I'd love to have one.

Thanks.

Matt

[Simon]

As someone pointed out, there are only 4 more that are likely to be
available within the next few weeks.
If people would be happy with an older model - the G3 or G4 variants -
then there maybe some of those as well. I will check when back next
week.

Martin Klang mentioned some older machines as well - so if you are
happy with something that is 5 or 6 years old instead of 4 years
please ask Martin.


I will also check what other surplus kit I can bring along for the
space


Simon

On Jun 2, 12:45 pm, "Matthew & Anna Israelsohn"

<misraels...@gmail.com> wrote:
> If there is a HP DL360 G5 available, I'd love to have one.
>
> Thanks.
>
> Matt
>

[tgreer]

Thanks, I'd only be looking at the G5s to match my current kit. If
they're taken then not a problem :)

[Nin Lil'izi]

I've been hankering for a server for some time and have always missed the boat so far.
I'm working on projects centered around supporting eastern european roma to speak up for themselves and fight for a better deal. Anyway, won't go tl;dr on that. But it involves bringing them all into the digital age on a large scale.

But have a budget of next to nothing. Barely no personal income and currently supporting this out of my own pockets. And looking for a server I can load up on serious amounts of ram and hdd and co-locate.
I have a prefernce for a 1u.... Just because hosting a 1u appear cheaper than a physically larger unit.
As its a requirement to be involving webservices supporting users in the 100k upwards bracket. (not all concurrently of course... but concurrency will be very high)... I really need somethign that can take a minimum of 16GB RAM...... Preferably more if possible, but as much as I'd drool all over 64gb in one place I realise that beggers cannot be choosers. But something that would support future upgrades if money can be found would be great
And something that I can later add a minimum of 4x 1TB HDD's.... Sata, because its cheaper.
Obviously, something with either cpu's that can supprt the above or sockets that could be upgraded for that purpose in the future.

I realise I am asking for ever so much and just edging on optimistic. I would of course be delighted with what was actually reasonably forthcoming
:)

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622

[Adrian Godwin]

Talking to Simon on Friday, it appeared there is a fair bit of kit to be moved.
I think there are already some volunteers to help - if there's not
plenty of cars available I can do a trip or two.

-adrian

[Jasper Wallace]

On Sat, 16 Jun 2012, Nin Lil'izi wrote:

> I've been hankering for a server for some time and have always missed the boat so far.
> I'm working on projects centered around supporting eastern european roma to speak up for themselves and fight for a better
> deal. Anyway, won't go tl;dr on that. But it involves bringing them all into the digital age on a large scale.
>
> But have a budget of next to nothing. Barely no personal income and currently supporting this out of my own pockets. And
> looking for a server I can load up on serious amounts of ram and hdd and co-locate.
> I have a prefernce for a 1u.... Just because hosting a 1u appear cheaper than a physically larger unit.
> As its a requirement to be involving webservices supporting users in the 100k upwards bracket. (not all concurrently of
> course... but concurrency will be very high)...

Any idea what kind of webservices? webmail? forums? different services
will have different loads and so need more or less cpu's or disks.

If you're only doing forums and mail then you probably won't be needing
large quantities of disk space. but if you want to store large quantities
of photos and video then thats another story

> I really need somethign that can take a minimum of 16GB RAM......

Pretty much anything you'l be able to get will be able to do 16Gb eaisly.

> Preferably more if possible, but as much as I'd drool all over 64gb in one place I realise that beggers cannot be choosers.
> But something that would support future upgrades if money can be found would be great
> And something that I can later add a minimum of 4x 1TB HDD's.... Sata, because its cheaper.

Disks are going to be the problem with a 1U machine.

It's better form a performance and reliability point of view to have many
small disks rather than a few large disks, the HP DL360 G5's take 6
2.5in SAS drives, but i don't think we can get 1Tb ones (you'd have to
buy them), and as you say, sata is cheaper...

This might be a good addition:

http://www.ebay.co.uk/itm/HP-MSA20-12x-3-5-SATA-HDD-Storage-Drive-Array-Enclosure-JBOD-RAID-SAN-NAS-/110889912879?pt=UK_Computing_Network_Storage_Disk_Arrays&hash=item19d18dca2f

> Obviously, something with either cpu's that can supprt the above or
> sockets that could be upgraded for that purpose in the future.
>
> I realise I am asking for ever so much and just edging on optimistic. I
> would of course be delighted with what was actually reasonably
> forthcoming :)

I think a HP DL360 G5 with as much ram and the biggest disks we can find
will be a good start.

--
[http://pointless.net/] [0x2ECA0975]

[Nin Lil'izi]

> Any idea what kind of webservices? webmail? forums? different services 
>

 will have different loads and so need more or less cpu's or disks.

First stage of the project is some bespokish social networking application. With us being a nation without territory. we were trying to build around a sort of virtual nation theme. Clever marketing maybe. But more of a hook to build the sense of (abeit nationless) identity to the roma nation amongst the younger generation and entice more of them to become active in fighting to change the poor deal they get in life and to defend their unique culture.

Why not facebook?
Because it's far too easy to mine. With most living under regimes of institutionalised racism with government sponsered gangs of door to door racists. This is very sensitive data. And the whole setup needs to be locked down like a ducks ass after somebody bought a new welder. Also, the other big problem is language. AFAIK, the first language of most Roma have never been localised in any software ever. At least not on the internet or anywhere it counts. So, everything produced would need a unique set of localizations.
A large number are totally illiterate. So ontop of people wanting to upload photo's (User videos some day in the future. But understand the technical challenges here would make this impractical for the forseeable future.) . Any informative content also needs to be available as either audio or video.
The next stage involves once people are getting online and interacting introducing demographic elements. So there is a way of guaging what the people want on a large scale. And polling for opinions on various issues which affect directly. So, that the NGO organisations that represent the Roma on the world stage have access to actual representative information about what the people want and think. And such data would also provide a spotlight to show up the organisations which are using Roma as political bargaining chips rather than being helpful. And strengthen the ones that are doing it right.

This project is not being undertaken by or the property of any organisation. It is intended to be run as an entirely politically and organisationally neutral entity of it's own. Only opening up polling and collating of data to pro-roma NGO's under a strict set of guidelines (that are still to be drawn up).
And the project is being worked on by a small group of individuals of Roma heritage. Mainly old skool tribal eldery types, with the help of a few experts in the other relevant fields that this crosses over into. For some reason this appointments me as queen of all things tech, but ho hum :)

So, don't need multi TB from the very start.... But room for storage redundancy and the initially tricky to guage initial growth without worrying /too much/ is where my mind at.
The only tricky thing that is priority in my mind. Is the high level of user concurrency. Which is mainly RAM + a godly level of dead reckoning.
I am considering architecting a few of the backend services as separate VM's on the iron from the start.... So that if it comes to expanding and breaking out into extra machines and devices in the future. Then it is an easy and straightforward task.

There is a plan to be applying to places for grants once this gets underway... But the initial setup and launch is coming out of our own threadbare pockets. Yes, I realise how insanely ambitious this is. But where's the fun in life without insurmountable challenges?

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622

 .
..:

[Mark Steward]

On Sun, Jun 17, 2012 at 12:50 PM, Nin Lil'izi <nin-l...@phoenixhaven.net> wrote:

> Any idea what kind of webservices? webmail? forums? different services 
>
 will have different loads and so need more or less cpu's or disks.

First stage of the project is some bespokish social networking application. With us being a nation without territory. we were trying to build around a sort of virtual nation theme. Clever marketing maybe. But more of a hook to build the sense of (abeit nationless) identity to the roma nation amongst the younger generation and entice more of them to become active in fighting to change the poor deal they get in life and to defend their unique culture.

Why not facebook?
Because it's far too easy to mine. With most living under regimes of institutionalised racism with government sponsered gangs of door to door racists. This is very sensitive data. And the whole setup needs to be locked down like a ducks ass after somebody bought a new welder. Also, the other big problem is language. AFAIK, the first language of most Roma have never been localised in any software ever. At least not on the internet or anywhere it counts. So, everything produced would need a unique set of localizations.
A large number are totally illiterate. So ontop of people wanting to upload photo's (User videos some day in the future. But understand the technical challenges here would make this impractical for the forseeable future.) . Any informative content also needs to be available as either audio or video.
The next stage involves once people are getting online and interacting introducing demographic elements. So there is a way of guaging what the people want on a large scale. And polling for opinions on various issues which affect directly. So, that the NGO organisations that represent the Roma on the world stage have access to actual representative information about what the people want and think. And such data would also provide a spotlight to show up the organisations which are using Roma as political bargaining chips rather than being helpful. And strengthen the ones that are doing it right.

This project is not being undertaken by or the property of any organisation. It is intended to be run as an entirely politically and organisationally neutral entity of it's own. Only opening up polling and collating of data to pro-roma NGO's under a strict set of guidelines (that are still to be drawn up).
And the project is being worked on by a small group of individuals of Roma heritage. Mainly old skool tribal eldery types, with the help of a few experts in the other relevant fields that this crosses over into. For some reason this appointments me as queen of all things tech, but ho hum :)

So, don't need multi TB from the very start.... But room for storage redundancy and the initially tricky to guage initial growth without worrying /too much/ is where my mind at.
The only tricky thing that is priority in my mind. Is the high level of user concurrency. Which is mainly RAM + a godly level of dead reckoning.
I am considering architecting a few of the backend services as separate VM's on the iron from the start.... So that if it comes to expanding and breaking out into extra machines and devices in the future. Then it is an easy and straightforward task.

There is a plan to be applying to places for grants once this gets underway... But the initial setup and launch is coming out of our own threadbare pockets. Yes, I realise how insanely ambitious this is. But where's the fun in life without insurmountable challenges?

My advice for this is: don't worry.  It's far more work to manage the code and content than the infrastructure, and that's what will determine whether you succeed in the first place.  A powerful server is only useful if you're serving vast amounts of data, or reselling it to people who want their own (reasonably powerful) servers.

Start off with a small server that will run whatever software you need, and let the provider worry about redundancy, networking and disaster recovery, so you can focus on the important things like keeping the setup secure.  There's no point over provisioning - stuff gets cheaper all the time, and your money is better spent when you're reasonably sure you'll need it.

Mark

P.S. not sure it changes anything, but there does seem to be a (very small) Wikipedia at http://rmy.wikipedia.org

[Nin Lil'izi]

Unfortunately. It's being pretty heavily stipulated. That super locked down dedicated hardware is non-negotiable.
Peoples fear is of the level. That there willingness to use any system would be more dependant on assurances of the utmost care and attention being taken to the security of the operation then anything else. Ok, broken content would be a bad thing too.... But no matter how great the content or representation going on... If anybody thinks the effort made for their security is half assesed, they won't touch it. And operating under siege conditions at times is expected. The democratic part of the project being the most important. Is what is particularly tricky. As I understand it there are a whole host national and international laws rocking against each other that dictate to a great extent how this data can be managed.
And while having people out there in the various countries assisting communities with getting online and voicing their oppinions is already pretty much taken care of when the tech happens. The concurrency required to deal the 100k's being polled within a week is going to eat RAM. And a co-located box with lots of ram... Is a lot cheaper to run per month then renting one of a similar spec.
I am looking into what degree I can leverage a 3rd party provider such as cloudflare (for teh obvious example) for reverse proxying the non-sensitive bukly stuff like images, videos, etc. And non-apache lightweight webserver alternatives.
Infrastructure and managing the tech side of the project is my major domain however. Am going to be handing off some chunks of web coding to a handy minion I'm being provided and hearing all sorts of awesome things about this persons abilities in the web design domain. And content seems to have no shortage of volunteer's to keep my inbox filled with useful stuff.

Ideally, once stuff is happening. I'll be wriggling as much as I can manage to keep my duties mostly limited to polishing the infrastructure. Admin stuff and train up other enthusiastic victims for moderatrix and content management jobs.

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[David Murphy]

 

I'm also curious how you're going to get 100K online right away. even online communities for fairly tech savy groups like uni forums often attact far less than 1% of their target group.

You'll have to moderate it to a certain extent since your upstream ISP can take issue with you if your users post certain forms of illegal content and it isn't removed promptly.
On the note of illegal content if you're running a service which includes image hosting or similar you may find you have to occasionally turn members details over to the police in order to avail of harbour laws related to certain forms of illegal content like CP to avoid getting in trouble yourself which while reasonable could cause mistrust of your service for fraternizing with the enemy.

Any plans for how it will be funded if it takes off as you hope? bandwidth for those kinds of numbers isn't cheap. Add revenue? it has the problem that google or whoever else serves the adds gets access to a lot of info about your users.

If security and anonymity is a really big issue you'd probably be better off setting your service up as a .onion Tor hidden service though that has it's own issues. 

[Martin Klang]

There are hosting providers who take a very clear stand on security, anonymity and anti-privacy laws.
Within the political activist sphere [1] [2] there's quite a bit of hands-on experience with these issues, including FBI and interpol interventions, server seizures et c.
There's also a cross-group initiative to promote more widespread adherence to user privacy [3].

Regarding Nin's project, you should really consider doing one thing, and do that well. Hosting/infrastructure is a different game from building a social network.


best,

/m

[1] https://aktivix.org/
[2] https://help.riseup.net/en
[3] https://and.nothingtohide.is/

[Nin Lil'izi]

The plan is. That there are people in the various countries. That will be registering people and getting them involved in person. And the numbers are for the number of people being polled for organisational and political reasons. Which as I said. Will have people on the ground dealing with communities directly.
People only deal word of mouth... Direct one to the other... Something only becomes fact when it's been told by somebody trusted and known directly... No other forms of communication have value and are disregarded.... So everything will involve people on the ground.... If it were not received by chinese whisper it is not true.... Digitizing the whisper chain is an interesting step.
I doubt that illegal content would be much of an issue, if at all. There will not be open signup's. And users will be strictly by invitation only, to only known and verifiable members of the communities. Things like CP, and other such illegal interests just do not exist amongst Roma... For cultural reasons... Those are gorger problems. It's one of those things that would be so unlikely, that the suggestion warrents a headdesk :3 ^_^
Things here work very differently then you'd expect from any other population. Mostly down to a complicated bunch of unique cultural stuff.
And I hope don't it doesn't take off. I really don't... But I'd never be that lucky :(. And knowing the culture and people involved. My bigger concern is that when they say they are going to poll 100k people in a certain week... They are probably underestimating the actual load that will be produced.
But rather then something being produced to hoist off on people hoping they take the bait.... This is a project that the target audience themselves actually got together and asked for. I just have too much of a big heart to say no... Even though I realise the extent of what is being asked for.
The financial problem is more worrying... I'm being asked to provide cost estimates for seeking funding... And finding that a bigger sodd.
My current stance... Is I am pushing to not start registering and teaching so many people to use it all in one go and start much slower... Of course, explaining the limits of technology where it's not understood and getting the message across that it's going to have to force people to work much slower and do a lot less then they traditionally burn through is as much of a challenge as the tech.
I'd like to get something running.... so I can realistically measure its footprint and then calculate costs based on that. With a controlled userbase. Also, the more complete and active a system it is.... The easier it will be to secure funding. There are large pots of money set aside doing nothing for pro-roma initiatives.... But it has to prove to be working, and actually producing the pro-democratic results promised. As the long goal is gaining sufficient political representation to enable an NGO to gain associate membership of the UN.
The social networking functionality really is just a sideshow on the edge of the actual aims and goals. And helping people get a uptill now totally absent voice to represent themselves politically is the actual game. And being able to have the 100k users log in once to vote on an issue and then bugger off again within the space of a week, is the core functionality that needs to be supported... Everything else is frosting.
I realise.... I suck real hard at explaining things.... But I'm a hackspacer... So should be expected.

tl;dr It sounds lofty and over-ambitious... will be technical challege... I realise this... But the stakes are high enough, it would be foolish to not give it the best stab I can.... Giving the silent a voice = priority.... social stuff = frosting... kinda intertwined... And I'll either succeed or you'll find me dead, slumped over a charred server.

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[Nin Lil'izi]

Oh, and I am definitly not dismissing your points.
They are all very obvious and very valid.
:)

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[Nin Lil'izi]

Oh, I realise this.
I am deliberately not detailing all the details of what the project entails. Because this is a public list.
Your just getting badly explained major points.
My plans for where to actually physically host this and arrangements over bandwidth and such... And relatationships with other politically focused organisations to avoid re-inventing the wheel. Are all in mind. Just prefering to keep off the list.
My game is hosting/infrastructure... I suck at building websites... terribly.... Which makes me glad I have a web monkey that knows what its doing.
Was just explaining more about what was going on.... Obviously... I am not trying to do all that stuff singlehandedly.
Just some people/project management and infrastructure are directly occupying my brain.

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[David Murphy]

it's good that it's invite only though on a non-anon or invite only service I wasn't thinking so much of people intentionally abusing the system but rather problems like 16/17 year olds sticking pictures of themselves up or for a distant boyfriend/husband and if they consider it a private service or private way of communicating I'd be surprised given teenagers being teenagers if it didn't come up now and then.
because people aren't doing anything actually wrong or hurting anyone they are less likely to worry about it.

given the law as it is though it can get the service provider ,you, into a lot of hot water unless you follow very exact legal steps which can require you to turn over details of the poster.

"The financial problem is more worrying... I'm being asked to provide cost estimates for seeking funding... And finding that a bigger sodd."

Te rule of thumb I was taught: if you're trying to build something large, stale, secure and complex.... make a reasonable guess at cost after adding up hardware, software and maintainance.... then multiply by 10 or more unless you've had a decade working in the area of large software project planning. treat that as your bare minimum, it'll probably cost more.

[David Murphy]

*stable, not stale

[David Murphy]

I like the idea but thinking about it some more... if the goal involves democratic voting of some kind that has to be of a quality that would satisfy UN observers..... you might have a problem.

e-voting has had a lot of problems and a username/pass on a site generally doesn't cut it. for important things like actual voting you'd need 2 or more factor authentication which would be a pain for the non-techies.

if grandma doesn't get or trust computers how do you prove that her kids aren't using her login to vote on her behalf for things she might not agree with?

Will the service only be for adults or for kids too? if kids abuse the service somehow like teens trying to hack into it how do you handle it? do they lose access and thus their vote forever? it gets even harder if outsiders harvest usernames and passwords and use the stolen credentials for maleficence. most laptops are infected with some form of spyware or botnet which could harvest their usernames/passwords and if re-registration or similar has to be handled face to face then it could be a significant headache.

[Nin Lil'izi]

I don't have all the answers regarding the voting.
There are specific university educated boffins who've mastered in the legal and political stuff that is going on, each specific to their own mother countries.
They're doing all the research and diplomacy... And pretty much handing me a list of requirements. And synchonized dances I must perform while chanting elvish over the consecrated toenails of Linus himself.

This is why, more than any other reason. I need dedicated, well specced hardware to host the eventual solution upon. And need to be spending my time worrying about getting the infrastructure right to an autistic level of perfection. Then what the mostly tech illiterate visionaries are dreaming up next... Ok, have to give some thought to their desires.... But in regards to UN related aspirations. Obvious priorities are obvious. And also why I can't be openly discussing in more than fuzzy terms, or the exact details of what is being planned in respect to many aspects on a public list.

But, it's kewl. People know what their upto... And fortunately their knowledge of what needs to happen is infinitely more competent then my ability to explain myself on a list :)

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[David Murphy]

bandwidth first.

100K people needing to vote in a week. lets assume that half of them leave it till the last day.

50,000 in a day

lowball: 10 pages each for guestimate.

call it 200kb average per page with images and html etc.

Will people have to download documents? images? video? upload/download?

furthermore they're not going to be perfectly spread out over those 24 hours. most will be using it during a 6 ish hour slot when it's evening in europe.

so lets assume that you may need to handle 30K people voting in a 6 hour slot.

(2mb*30,000)/6

so call it 10GB per hour (lowball) or 3mb per second upstream, more if you don't want people to click and have to wait for 2 mins which is very much a buisness connection with a reasonable fraction of that downstream.

I can't guess how much CPU power you'd need without knowing how much you want to do for each user but at the moment I'm assuming you just want to let people log in,
read half dozen small pages or look at a few small images, vote then log out rather than view hundreds of large pages or images like facebook or submit and read lots of forum posts.

you'll need a database backend, probably on it's own server which can handle that kind of traffic though that depends largely on your implementation and database schema.

if reliability is important you'd need a duplicate web server and database server for failover in case of the server dying and prossibly more for other single points of failure.

you don't want to lose data so you'd need to budget for a decent commercial grade raid array and probably long term tape backup to be stored at a second site.

budget in some maintainance.

the software... the software could be fun depending on how complex this all is and could either be reasonably straightforward or a horrorshow to make reliable and secure. really secure isn't easy.

do you need it to be secure from kids or secure from governments or organisations with money? if the latter you'll have a tough time even with a very simple site with very few features.

[tgreer]

I think we may have went a little OT here...

[David Murphy]

you could hack something together for a tiny fraction of that that would work great with a thousand users and possibly, maybe sort of work for the kinds of numbers he wants.

it all about how reliable, secure, fast etc he actually wants it to be.

It's a common sin amongst techies that we tend to underspecify.

 

[Nin Lil'izi]

Thanx. I have done these sorts of calculations.
Went outside to scream at the sky for mercy. Then came back inside and did them again.
I realisticly expect upon expected conditions at startup... That the system would have to support 2-3k people per hour. To fill in a registration form... view a page.... then submit something.

It's the demands upon the hardware... That I'm feeding back in the other direction. That the people who are assisting with the registration and voting in person. Stagger their processing of people over a wide and even space of time as possible.
Most people simply can't just register and vote on their own behalf without help. Actually the majority can't ... Most are barely literate. Literacy in places is seen as an uncommon and high value skill that has very real bargaining power. It's just not the given that it appears to be in some parts of england I've been. Hell, I live in a community of 30 people just outside london and only a small handful of us can read
It's another reason why large scale democracy has been impossible so far.
People will have to go out physically and help people register and teach them how to vote in the first place. At the start of exploring this project. I did not understand what voting was. And what the point of it is. Been a learning experience so far.
So, to start with... On that level... Getting peoples voting to be relatively evenly spaced is going to be vageuly achivable... Because it's a case of whoever is on the ground floor only being able to register and tutor so many people at a time.... And they'll be following such instructions... And they're registrations and assistence will be on the timetable of the people helping.
I'd like a seperate db server.... who wouldn't? ... Unless one falls out the sky and leaves a firey hole of death and destruction through the roof of my caravan. I don't fancy my chances of that happening soon enough. So, just going to engineer the components as seperate vm's ... So when expanding becomes necasary and hardware appears.. breaking it out into respective places is an easy task. I'd like to add a reverse proxy so I don't waste cpu cycles and ram serving the static content items where the dynamic content generation is happening too. I think the forward proxy is the place where the SSL and possibly IDS, given the need to be aware of hostile government interests. needs to be offloaded too aswell. But it's a question of where another additional device will come from. I'm also exploring handing off some of the non-sensitive static assets to a cdn. to help minimise load. It feels just rude to start asking for multiple servers..... I realise I do eventually need multiple servers for this to work.... But a single big piece of iron carved up and ready to expand when needed is a good starting point. I'd personally rather have a thing to show off, that works and is in a form fit for scalling... Before trying to acquire all the additional hardware to do the actual scaling.
It won't be perfect... it will probably be quirky and slow as hell. But I'd still rather make my best effort to produce /something/ .... Rather than just let things remain with there currently being nothing. And many people across the world not even realising a something is possible.
I really am counting on that once a something is working and run a couple of rounds. That funding for a more appropriate level of infrastructure to support it will be successfull... But that's the job of people with fancy qualifications and educations. Totally out of the grasp of an uneducated geek like myself.

I keep reading this back... And it just sounds angry and combative... That's not how this is intended.... It's the only way I know to express myself and don't have a clue how to fix my language to sound more neutral :) .... This is intended pleasent chatty, really :)

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[Nin Lil'izi]

It's a she ;-)

And there are many variables involved.
Now, I'm a geek with a crazy project... That will either achive something kewl. Or die trying.
I posit that I'm definitly not a member of the 'herp derp, gonna make the next failbook. it'll be just like it but twice as awesome' crowd.
I have some pretty tightly architected ideas, that while not guarenteed instant success... Are interesting enough to put to the test.
And like the typical geek, my ability to communicate these things is non-existent compared to my ablilty to impliment them. And really, was just trying to share some of the things involved. As a background to what I'm upto. And how will be greatefully putting any donated server hardware towards a novel project under an unusual for a hackspacer set of circumstances.
Don't want to be the next failbook... or the next anything... or even big... Just don't want to remain the worlds solitary cyber gypsy and try and give some of my people the chance to climb up and join me on my geeky pedestal.

Regards,
Nin lil'izi

GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
 .
..:

[Alex Pounds]

On Mon, Jun 18, 2012 at 12:14:29PM +0100, Nin Lil'izi wrote:
> I doubt that illegal content would be much of an issue, if at all.

> users will be strictly by invitation only, to only known and verifiable
> members of the communities. Things like CP, and other such illegal
> interests just do not exist amongst Roma...

You're looking at this the wrong way. Your concern here is not "Will
anyone upload anything illegal or actionable?". Your concern should be
"Will any of my members be accused of uploading anything illegal or
actionable?". Regardless of whether *you* think you are hosting anything
pertinent to a civil or criminal case you may be faced with a lawful order
to disclose user data. I don't think it's quite the no-brainer you
suggest.

--
Alex Pounds
Web Developer & Photographer

http://alexpounds.com/ | http://ethicsgirls.com/

[Simon]

OK, there are now a number of servers stacked neatly under the desks
in the space

we have done an inventory of all the machines and I will be in
tomorrow evening and try to go through the emails etc to see who
wanted what - either that or I will put the inventory on the wiki and
then people can add their name and purpose to the machines, along with
any donations they would feel like making to hackspace


I would like to keep a couple of the bigger machines in the space to
play around with various virtualisation technologies and then to
settle on one, maybe with a view to using it as infrastructure to
provide people with virtual machines on demand - I have spoken with a
few people who think there could be a demand for such


In the meantime, please do not hack any of the machines or
components. I will write more details about the kit tomorrow


Simon

> /Nin lil'izi/

>
> GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
>  .
> ..:
>
> On 18/06/12 09:54, David Murphy wrote:
>
>
>
>
>
>
>
>
>
> > I'm also curious how you're going to get 100K online right away. even
> > online communities for fairly tech savy groups like uni forums often
> > attact far less than 1% of their target group.
> > You'll have to moderate it to a certain extent since your upstream ISP
> > can take issue with you if your users post certain forms of illegal
> > content and it isn't removed promptly.
> > On the note of illegal content if you're running a service which
> > includes image hosting or similar you may find you have to
> > occasionally turn members details over to the police in order to avail
> > of harbour laws related to certain forms of illegal content like CP to
> > avoid getting in trouble yourself which while reasonable could cause
> > mistrust of your service for fraternizing with the enemy.
> > Any plans for how it will be funded if it takes off as you hope?
> > bandwidth for those kinds of numbers isn't cheap. Add revenue? it has
> > the problem that google or whoever else serves the adds gets access to
> > a lot of info about your users.
> > If security and anonymity is a really big issue you'd probably be
> > better off setting your service up as a .onion Tor hidden service
> > though that has it's own issues.
>
> > On Sun, Jun 17, 2012 at 2:11 PM, Nin Lil'izi

> > <nin-lil-...@phoenixhaven.net <mailto:nin-lil-...@phoenixhaven.net>>

> >     /Nin lil'izi/

>
> >     GPG Fingerprint: C510 909B 811E D6F5 0DFF 5D91 CF03 8FEA FD69 4622
> >      .
> >     ..:
>
> >     On 17/06/12 13:12, Mark Steward wrote:
> >>     On Sun, Jun 17, 2012 at 12:50 PM, Nin Lil'izi

> >>     <nin-lil-...@phoenixhaven.net
> >>     <mailto:nin-lil-...@phoenixhaven.net>> wrote:
>
> >>         *>* Any idea what kind of webservices? webmail? forums?
> >>         different services *>*

> >>          will have different loads and so need more or less cpu's or disks.
>
> >>         First stage of the project is some bespokish social networking application. With us being a nation without territory. we were trying to build around a sort of virtual nation theme. Clever marketing maybe. But more of a hook to build the sense of (abeit nationless) identity to the roma nation amongst the younger generation and entice more of them to become active in fighting to change the poor deal they get in life and to defend their unique culture.
>
> >>         Why not facebook?
> >>         Because it's far too easy to mine. With most living under regimes of institutionalised racism with government sponsered gangs of door to door racists. This is very sensitive data. And the whole setup needs to be locked down like a ducks ass after somebody bought a new welder. Also, the other big problem is language. AFAIK, the first language of most Roma have never been localised in any software ever. At least not on the internet or anywhere it counts. So, everything produced would need a unique set of localizations.
> >>         A large number are totally illiterate. So ontop of people wanting to upload photo's (User videos some day in the future. But understand the technical challenges here would make this impractical for the forseeable future.) . Any informative content also needs to be available as either audio or video.
> >>         The next stage involves once people are getting online and interacting introducing demographic elements. So there is a way of guaging what the people want on a large scale. And polling for opinions on various issues which affect directly. So, that the NGO organisations that represent the Roma on the world stage have access to actual representative information about what the people
>

> ...
>
> read more »
>
>  signature.asc
> < 1KViewDownload

[William Hay]

On 19 June 2012 01:28, Simon <skl...@gmail.com> wrote:
> I would like to keep a couple of the bigger machines in the space to
> play around with various virtualisation technologies and then to
> settle on one, maybe with a view to using it as infrastructure to
> provide people with virtual machines on demand - I have spoken with a

If people want virtual machines it might be worth finding out how many
could make use
of/would prefer something lighter weight like LXC and having that as well.

William

[Simon]

I do like the idea of simple containers on a machine to keep some
things isolated but not everyone will want linux though and having
your own full VM to play with with whichever OS and OS variant you
want would probably be attractive to a number of people

[William Hay]

On 19 June 2012 18:18, Simon <skl...@gmail.com> wrote:
> I do like the idea of simple containers on a machine to keep some
> things isolated but not everyone will want linux though and having
> your own full VM to play with with whichever OS and OS variant you
> want would probably be attractive to a number of people

Which is why I used the phrase "as well". There are some applications that
don't play well with hardware virtualisation but do with something
lighter weight.

[Simon]

This round of servers has now been audited and a wiki page created
(Thanks Phil)
http://wiki.london.hackspace.org.uk/view/Super_Server_Bonanza

I am also proposing a loose bidding process for these machines - I
would like to see hackspace members benefit as much as possible.
I have linked another page for this bidding process from that - I am
proposing a 2 phase bidding process with the first phase being for
cool projects that might not otherwise happen and the second phase
being for some cash for the HackSpace coffers so if you just want a
machine then name your price.


I have also had a number of people email me and also there are many
messages in this thread but can I ask all who want machines to make
your wishes on the Wiki rather than here so that all the info is
simply correlated in one place


There is also a possibility of another round of similar machines later
in August/September,

Simon

[Peter "Sci" Turpin]

Wow, those are some impressive specs!