Trusted Testers ?

[bobololo]

Dear video enthusiasts,

As we're quite close from a new beta round of our avc encoder, we're this time faced to the testers selection issue. You all certainly know that we suffered from a leak during our previous beta round and that unfortunately has caused us to be more restrictive on the testers selection.

The purpose of this post is to collect your comments and opinions in order to setup a trusted beta testing environment.

A very first draft of what we could do would consist in those following rules :

- A subset of trusted testers (mostly testers from first beta) is selected among all volunteers.

- Then those initial testers could give their consent to other volunteers.

- To become a new tester, a volunteer needs to have the consent of at least two trusted testers.

- An initial trusted tester can give his consent to three volunteers at most.

- If a tester leaks some material he's obviously banned from the test, but testers who gave their consent to him are fired as well forcing people to give their approval to persons they seriously believe in.

The drawback I see in those rules is that newcomers don't have any chance to take part of the test. But I definitely don't have any clue of how we could trust a totally unknown volunteer (unless asking them their credit card number ;).

What do you think ? Do you believe these rules are fair enough ? Have you any idea regarding the inclusion of totally newcomer to the test ? All point of view and suggestions are welcome !

-- bobololo

[Sirber]

Seems cool. Hope it won't leak this time. Looking forward to the new beta tests! :D

[hroo772]

Even though I was not part of your test last time I understand your feelings about a leak. I would like to help your testing team out so that your product can get all the bugs out. Thanks.

[Joe Fenton]

I'm a little surprised you don't make folks sign an NDA. Email an NDA form to the tester. Have them print, sign, scan, and email it back. I've done that several times.

I noticed you didn't make any mention of the kind of systems you're looking for with testers. You might at least mention the minimum system to avoid the trouble of turning down people with 32M 586 90MHz systems. :D

[IgorC]

Rules are fair and clear.
Is it test of High Profile or a new version of main AVC?
What new features are coming out? subtitles, menu, new GUI.
Yes, it´s very hard to beleive to somebody. At least it´s possible
to ask for some personal information.

[acidsex]

I also support signing NDAs. Sorry but these days, its very hard to put my name and trust on the line for someone on the net I have never met before. NDAs are the only way to go for you all to have some legal binding recourse.

That said, I look forward to being a part of the next round of testing.

[Blue_MiSfit]

I would love to be a part of the new round, who trusts me? :)

[ChronoCross]

If you need a tester I can do it as well. Let me know.

[Mug Funky]

it would be intersting to take part in this test (if y'all trust me).

i have a few questions though:

what kind of sources are you into? i assume a wide variety (not just anime), also, what kind of resolution? D1 size? also (here's a big one) interlaced or progressive?

i ask because i have access to bucketloads of sources of all kinds, but i'd say 60% is either pure video or standards-converted stuff that is now interlaced. sources would be m2v's (don't have the time or disk-space for digibeta captures...).

also, the question about systems is a good one. i have a p3 733 with 512 of regular SDRAM.

only problem is i may not have enough time to do loads of testing.

[Neo Neko]

I am always up for some testing. I have several systems (though all AMD ones) avalible to run on. My only possible issue is I don't have a version of Nero 6 or recode. :( Got a version of 5 somewhere that came with my DVD burner. Not that it's either here nor there. :p Although I have been toying with the idea of getting the latest nero including recode when I have funds avalible in the future. ;)

[hhanh00]

Originally posted by Joe Fenton
I'm a little surprised you don't make folks sign an NDA. Email an NDA form to the tester. Have them print, sign, scan, and email it back. I've done that several times.

I noticed you didn't make any mention of the kind of systems you're looking for with testers. You might at least mention the minimum system to avoid the trouble of turning down people with 32M 586 90MHz systems. :D

Even with a NDA, if the company doesn't have the will or the means to take legal action, there isn't much protection against a leak. What about adding a watermark to each build and adding a timebomb? You guys probably already thought about it, right?

Thanks,
--h

[Koepi]

Guess how they found _who_ leaked the codec last time.

I guess I'm no trustworthy tester either, am I?

(Just kidding. I've no time to test another codec :) )

Cheers
Koepi

[dragongodz]

Email an NDA form to the tester. Have them print, sign, scan, and email it back.
hmm and if its say someone like me who does not have a scanner or access to one ? sorry an NDA doesnt always work and discounting someone because they are unable to sign one is selling some people short. i did some early testing for VP7 for example, did anyone even know that though ? no NDA was needed and i was still trustworthy.

Then those initial testers could give their consent to other volunteers.
thats better. my past contributions to the forum should show the type of person i am. maybey making a list of potential new testers and asking ,privatly, for opinions of these people from the trusted initial testers would be better than having to have the explicit consent of 2.

just thoughts anyway. :)

[Andrey]

Hi guys !
>>I also support signing NDAs. Sorry but these days, its very hard to
>>put my name and trust
NDA is good, but in many countries it will mean nothing in legal terms. :(
Idea of two trusted tester is interesting, but would it work on practice ? I do not want to give my (for example) word for people I do not know and saw 10 posts of them...
Need to think about it... :(

[hhanh00]

Well... I'd like to throw my hat into the ring. Anyone would like to sponsor me?

Thanks,
--h

[bobololo]

Originally posted by Joe Fenton
I'm a little surprised you don't make folks sign an NDA. Email an NDA form to the tester. Have them print, sign, scan, and email it back. I've done that several times.

I noticed you didn't make any mention of the kind of systems you're looking for with testers. You might at least mention the minimum system to avoid the trouble of turning down people with 32M 586 90MHz systems. :D

Even with a NDA signed, we have no way to prevent a malicious person to use a false identity & address. As long as we're dealing with a single person (not a company) it will be hard for us to verify whether he's cheating or not.

Regarding the requirements, they will come in time. This thread wasn't intended to give the kick off of a new beta but instead a discussion about how it could be done :)

[Latexxx]

Using dongles would be some serious protection and would force people to give their real addresses. On the other hand it would be awfully expensive to send dongles all over the world.

[acidsex]

Originally posted by Latexxx
Using dongles would be some serious protection and would force people to give their real addresses. On the other hand it would be awfully expensive to send dongles all over the world.


Not to mention that dongles are easily crackable. Just ask Avid.

I would support the 2 recommendation system however I will NOT recommend/ednorse anyone. As I said before, no way am I going to put my nameand integrity in someone else's hands. Call it paranoia, Ill just call it being safe and prevention and protection.

They could always hard code the testers real name, ss#, address, etc... so if they idiot decides to pirate it, then not only does Ateme/Nero lose something, the moron will have his info out there on the net for others to do him harm. Kharma's a bitch sometimes. If testers are responsible and dont share, then their personal information will NOT fall into the wrong hands.

[Teegedeck]

Reasonable setup.

As always I am dead curious for as long as the codec hasn't been released. After the release I don't really care about closed-source stuff. Thing is that a) I don't have much time, so if I were to participate I could only test performance on two or three clips, and b) I don't have Nero 6 so I could not compare your new codec to that version. So decide whether this is any help for you at all.

[yaz]

would be mottoed like 'trained watchdogs are warmly welcomed, some trust-paranoia is non-exclusive' ;)
in my very personal and very humble opinion, this way u wouldn't get anything else but inefficiency. if i were a trusted person i would follow the strategy; not to invite anyone else as it minimizes the risk of being banned in the end ...
... if that 'ban' is really sg which would be taken seriously. amof, i can't see any possible way of 'banning' which would be preventive (a/o deterrent) enough.
... aso aso ... such systems have many drawbacks and are only good for 'laying the blame on someone else'

what u need is 'personalizing' the testers. one way is contracting them personally. (no branching nor 'trusting the trustee' neither sg) the other (less reliable) way is 'personalizing' the codes released. each tester obtain a unique digitally signed copy. it'll provide only 'tracebility of leakage' but it'd be much more than that given by the system u outlined.

good testing
y

[Soulhunter]

Originally posted by bobololo

Even with a NDA signed, we have no way to prevent a malicious person to use a false identity & address. As long as we're dealing with a single person (not a company) it will be hard for us to verify whether he's cheating or not.

Dumb idea...

Why not demand a picture of the identity card ???


Bye

[bobololo]

Originally posted by yaz
what u need is 'personalizing' the testers. one way is contracting them personally. (no branching nor 'trusting the trustee' neither sg) the other (less reliable) way is 'personalizing' the codes released. each tester obtain a unique digitally signed copy. it'll provide only 'tracebility of leakage' but it'd be much more than that given by the system u outlined.

The testing materials in the previous beta *were* signed and didn't prevent from leak.

[bobololo]

Originally posted by Soulhunter
Dumb idea...

Why not demand a picture of the identity card ???
Ask Cruncher to show you his identity card ;)

[Sharktooth]

IMHO digital signature would be the best choice.
Each encoder get digitally signed with user data.
Users signs a NDA and if a leak happen you know who leaked the software and you can sue him/her.

[yaz]

Originally posted by bobololo
The testing materials in the previous beta *were* signed and didn't prevent from leak. must quote myselfOriginally posted by yaz
...it'll provide only 'tracebility of leakage' or in other wordsOriginally posted by Sharktooth
...if a leak happen you know who leaked the software and you can sue him yep ... personalize ... or have i mentioned it yet ? ;) (hmm, is it an english word at all?)

btw, sharky, why do u think girls wouldn't leak ? :)

the bests
y

[Sharktooth]

Doh!

[IgorC]

It´s possible to join documents as Passport, University card etc. Maybe it´s easy to falcificate 1 document, but not a few.

I can send scaned copy of passport. But wich one? :rolleyes:

[Sirber]

Please do NOT email or pm me about being a tester. I don't even know if I'm a trusted one. Thanks.
Originally posted by IgorC
It´s possible to join documents as Passport, University card etc. Maybe it´s easy to falcificate 1 document, but not a few.

I can send scaned copy of passport. But wich one? :rolleyes:
Too much troubles to manage...

[Razorblade2000]

well, sounds fair to me...
nothing to cover = nothing to hide :P

[Andrey]

>>Why not demand a picture of the identity card ???
Atheme can have legal issues with it, because storing personal data in many countries require to register the "storage" and the storage goal in a special goverment commission. At least, here it is.

[dragongodz]

Users signs a NDA and if a leak happen you know who leaked the software and you can sue him/her.
Why not demand a picture of the identity card ???
and i will quote myself in reply.
hmm and if its say someone like me who does not have a scanner or access to one ?
besides if you are so untrusting of a person you need photo i.d. etc then just dont use that person.

I would support the 2 recommendation system however I will NOT recommend/ednorse anyone. As I said before, no way am I going to put my nameand integrity in someone else's hands. Call it paranoia, Ill just call it being safe and prevention and protection.
if you are not willing to recommend anyone then you do not support it at all. would you give a general opinion of people on a list though if there was no recommending or come back against you ?

[Leak]

Originally posted by Koepi
I guess I'm no trustworthy tester either, am I?

Well, I guess I as well shouldn't even think about applying, considering my username... :D

(Not that I currently have the time for it, anyway... ;))

np: Autechre - Part 1 (XLTronic Radio 25.01.05)

[Selur]

if there's still room for testers I'd like to test too :D
(I like testing new stuff,...)

Cu Selur

[ChronoCross]

Suing someone for leaking a codec would be rather pointless. Waht are you gonna do garnish his wages for the next 100 years to try and make even a few thousand dollars? I think that's the main reason that things get leaked, people don't have anything that would be worth suing. It would cost nero more in legal fees than they would get out of the individual. Which is why their approach of trusted testers goes to the very core of our community. It shows how much each person trusts each other. I would definitely trust quite a few people on this forum and for those of you who have been around longer probably have more reasons to trust people.

Now if Only I knew who the trusted testers were and could get them to vouch for me lol.

[LigH]

Unfortunately, I don't own a scanner.

But I'd like to participate again (because I have a rather slow and lower-level CPU, you might be interested in speed and non-SSE compliance reports).

So if any trusted tester would like to give me one of his consent, I wouldn't mind.

[JoeBG]

It would be a priviledge to be a member of the testing team and working with you all :)

[bob0r]

try open source, problem solved

[Sharktooth]

LOL :D

[acidsex]

Originally posted by dragongodz
and i will quote myself in reply.

besides if you are so untrusting of a person you need photo i.d. etc then just dont use that person.


if you are not willing to recommend anyone then you do not support it at all. would you give a general opinion of people on a list though if there was no recommending or come back against you ?

When I said I support that 2 recommendations, its just that. I support the program and find nothing wrong with it. My choice NOT to recommend anyone thus placing my integrity on the line is my CHOICE. While others may not be paranoid of such things, I have a lot more at stake then most. Whereas a regular tester who works 40 hours a week couldnt possibly be sued and have major damages taken from them. In my situation, I have a lot of income that could be sued for and I am not willing to put my lively hood on the line for anyone. This is more than a beta test for some of us, especially when you figure in legal liabilities and financial ramifications.

[SeeMoreDigital]

All the usual suspects here I see!

I'm really pleased that we've been considered for another test, so I'm up for it... if I'm considered worthy :)


Cheers

[IgorC]

I think probability that beta will be leaked in second testing will be much minor than in first. Because second test will be integrated from 80-90% aprox. (I don´t know presice numbers) from old testers of the first test.
How much people leaked their beta in first test? Only one person (besides it was enough).
Probability of beta leak of second test will be depending of the new testers that forms 20-10% (aprox) from total number of testers. But the each new tester has to have 2 votes. It´s enough to prevent bad results.

[Sagittaire]

[Mode joke On]
it's very simple ... don't believe Russian tester ... lol
[/Mode joke Off]

[Sirber]

Bad jokes...

[Teegedeck]

When criticising this whole approach don't forget that our friends at Ateme inevitably had to come up with some kind of security scheme as a precondition if they were to justify getting their beta-testers from this community once more. What they came up with is as good and non-formal as any solution based on trust goes. Time to let it go...

[Leo 69]

To my mind, the probability of leakage is close to zero, just because
th same people will test the codec. And this... voting system is a bit weird. It evokes some not pleasant feelings about the whole situation.
But special measures seem to be necessary sometimes, yeah..

[Teegedeck]

It is not so unusual. Very basic and important a concept, as a matter of fact. Google a bit on 'webs of trust' and 'PGP/GPG'.

[Leo 69]

Well I'm actually very honest. :D I would be in seventh heaven if I'm in :)

[Sirber]

Web is a place where it's easy to lie and hide :)

[LigH]

I don't mind some level of security. For such a level of technological advantage, it appears quite logical to me to ensure that.

I just wonder where the "root" of trusted testers start, who belongs to the "hard core". And: What other testers may get in mind to advertise for recommendations...

[BBugsBunny]

I'd like to participate as well.
I probably have got a rather rare system - a Dual Xeon 3.6 GHz system (EMT 64 ready).
I haven't participated in the first testing round, so I don't know if I will have a chance to get into the second but for sure I would like to...