IACR Publication Reform :  Cryptology ePrint Archive Forum
Discussion related to IACR's current and future publications: conference proceedings, Journal of Cryptology, and revolution of IACR's publications.  
Goto Thread: PreviousNext
Goto: Forum ListMessage ListNew TopicSearchLog In
IACR publication reform - my two cents
Posted by: hoerder (IP Logged)
Date: 30 May 2013 07:03

Unfortunately I wasn't able to attend EC yesterday so apologies if some of the points have been raised already.

It seems to me that not including the workshops in the Journal scheme is contra productive as you'll have to review journal rejects again for those conferences. Furthermore, those conferences will get journal rejects only -- journal acceptances that do not get into C/EC/AC are out of the pool. However:
1) It may be reasonable to have a one or two year delay for bringing the workshops into the journal system just to make sure that the system works well enough and not to have all eggs in one basket immediately.
2) It may be reasonable to combine the workshops with regular, co-located post-grad (summer) schools on en vogue topics. If everyone knows that IACR organizes 2 to 4 summer schools on state of the art topics every year and if there's an inofficial price limit on those summer schools, then everyone can include a suitable multiple of that limit in grant applications for PhD studentships and you'll never lack people attending those summer schools & workshops.

If you want a quick turn-around, include a "letters to the editor" or short paper section. Sometimes a paper triggers an idea that should be published but isn't substantial enough to make a full, high quality paper. Instead of padding it with background blubber and sending it to a crappy conference or keeping silent until you've accumulated enough substance for a full paper, people could just send a short, down-to-the-point paper to the journal and get it published fast.

The journal should have a good categories/keywords/tag system to help with lit research. Coherent tagging rules would be helpful and reviewers should make sure that the paper has been tagged properly (in case of acceptances). Right now, when I submit a paper I specify all keywords that seem relevant at the time of submission. With re-submissions, I noticed that I never come up with the exactly same list of keywords (unless I look up what I used the last time). I take that as indication that my intuitive ad-hoc approach to keywords probably does not help a lot. (Keyword based search should beat googling in my opinion.)

Finally: What are you going to do with the backlog of papers that are already in the submission-resubmission cycle when the Journal starts. (I don't know how large that backlog is/will be.) If it's just a few, I don't think that it's going to be a problem but if there is a backlog of a few hundred papers you risk that the journal either starts pear-shaped or that reviewers initially are too tough to get rid of the backlog.

Cheers,
Simon Hoerder

Re: IACR publication reform - my two cents / workshop problematic
Posted by: cbw (IP Logged)
Date: 30 May 2013 10:58

Hello,

I agree that there is a foreseeable problem with the workshops - both IACR and non-IACR. In a nutshell, we will "vamperize" them with the "Proceedings of the IACR" (proc.IACR)and suck out of them all the good publications.

I don't think that the workshops will disappear. We'll get either an informal solution or a formal solution (or a mix of both):

(1) Informal solution: Workshops will accept any talk where an author says "look, I have an IACR Publication. Would you like me to speak at your workshop?" Out of sheer lack of submissions, workshops will agree. Alternatively, the will actively "hunt" publications in proc.IACR that have not been accepted for AC/EC/Crypto-Presentation. Both information is public, so we cannot prevent this.

(2) Formal solution: Workshops get (somehow) incorporated into proc.IACR and agree among each other which presentation goes to which workshop. The system will be self-regulating: Authors that thing their article is too good for workshop X will decline an invitation and hope to get invited for workshop Y. If their article is indeed good enough, this may happen. And it will be difficult to manage in the beginning.

At least for the IACR-workshops (CHES, FSE, PKC, TCC), I would like to see an offer from proc.IACT towards them. If they don't take it, that's fine. However, there should be a genuine offer. I agree it should not be in the first year of proc.IACR - things need to fall into place first. But from year two onwards, they should be included.

Secondly, I would like to extend this offer to all workshops that have "in cooperation with" status - see [www.iacr.org]. Again - it should be an offer, nothing more.

Including more workshops could prove difficult. In particular: If a workshop does not even fulfil the rather minimal criteria of ICW - why should we include it into proc.IACR?

So at the end, there will be some workshops that are formally associated with proc.IACR (maybe on a year-by-year basis) and other workshops that can informally draw from the publications there to increase the number of reviewed talks they have. How they fill the other slots is a different ball game and I am very interested to see different solutions implemented :-)

Best,
Christopher

Re: IACR publication reform - my two cents
Posted by: cbw (IP Logged)
Date: 30 May 2013 11:04

Hello,

Simon Hoerder wrote:
-------------------------------------------------------
> Finally: What are you going to do with the backlog
> of papers that are already in the
> submission-resubmission cycle when the Journal
> starts. (I don't know how large that backlog
> is/will be.) If it's just a few, I don't think
> that it's going to be a problem but if there is a
> backlog of a few hundred papers you risk that the
> journal either starts pear-shaped or that
> reviewers initially are too tough to get rid of
> the backlog.


Thanks for raising the question. To my knowledge, it has not been addressed yet. But I have not been in Athens either ;-)

I *does* have to be addresses - but it's not really difficult: We have an online-journal now, so there is no bandwidth limitation anymore. So in 2016, we can already review the three flagship conferences using the following grades:
(1) Accept
(2) Reject
(3) proc.IACR

All authors with (3) get asked if they want to submit to proc.IACR. If they do (presumable with small changes), we can already publish the first article in proc.IACR in May 2016 :-) And as I said: There is no limit. If we accept 300 articles within the first 6 month, that's fine.

So I guess the backlog will not add too much to the overall workload.

Best,
Christopher



Please log in for posting a message. Only registered users may post in this forum.