Added TVCG/IEEE Vis #77
Conversation
IEEE TVCG-VIS/VR should be included in the publication list, since IEEE VIS and IEEE VR are large specialty conferences that each have about 1,000 and 500 attendees annually and their proceedings are published as special issues of IEEE TVCG. There are more than 50 top institutions around the world that have published regularly in IEEE VIS and IEEE VR with IEEE TVCG since 2000. This arrangement is identical to ACM TOG/SIGGRAPH, which you’ve listed under Computer Graphics. The CS attendance and participation level at these conferences surpass many system/AI/theory conferences listed under csrankings.org.
|
No journals are currently included except those that are the publication venue for conferences. Attendance is not a criterion for inclusion. |
|
On 18/08/16 16:18 , Emery Berger wrote:
PS I am answering directly by email because the thread on github for |
|
I support this pull request. IEEE VIS and the relationship to TVCG are equivalent to the relationship of SIGGRAPH and TOG. And Visualization is a very active and growing CS research field. |
|
I also agree that missing IEEEVis excludes a number of active CS researchers. IEEE Vis (http://dblp.uni-trier.de/db/conf/visualization/index.html) has existed since 1990 and merged with InfoVis in 1995. Paolo's pull request is probably the easiest way to include this. As mentioned, articles submitted to TVCG are frequently presented at IEEE Vis, and we currently view the conference as equivalent to journal. This relationship has existed since 2006 or so, but was formalized in 2011, see: https://www.computer.org/web/tvcg/conference-partners. A nice overview of this history is here: http://www.vispubdata.org/site/vispubdata/ (scroll down to "Notes on IEEE VIS, its Child Conferences and Naming") You would likely also want to InfoVis (http://dblp.uni-trier.de/db/conf/infovis/index.html) and VAST (http://dblp.uni-trier.de/db/conf/ieeevast/index.html), for the complete history of everything that we call the current IEEE Visualization conference. If you want to get more precise, you could add only the above three conferences and then subselect only those issues of TVCG that corresponded to proceedings from the conference. This would be representative of the current state of affairs, where in 2010 all three were finally called conferences under the 2006-flagship IEEE VisWeek (later named to IEEE Visualization). Our relationship to TVCG is a bit more subtle than SIGGRAPH / ToG, as TVCG also host special issues from other Vis conferences (e.g. Pacific Vis) as well as non-vis conferences (e.g. IEEE VR, ISMAR, etc.). Similarly, if you wanted to add VR (e.g as the "third" graphics conference), you would add both the VR conference (http://dblp.uni-trier.de/db/conf/vr/index.html) and only its related TVCG issues. Although, a Graphics, Vis, or VR person might argue that these are three different research areas altogether. Happy to help out mucking up csrankings.py, if you'd like. |
|
IEEE TVCG is basically a publication venue for IEEE vgTC leading conferences, like VIS and VR -- just like ACM TOG is for SIGGRAPH and SIGGRAPH Asia (plus other reg. submissions). So, if you list TOG, which is a journal, then IEEE TVCG should be listed. IEEE TVCG publications are also presented at the top vgTC conferences, just like TOG papers. The H-5 index (a measure of impact) of TVCG is higher than those for APLOS, ISCA, Micro, MobiComp, SIGMETRICS, IJCAI, AAAI, EMNLP, NAACL, CRYPTO, EuroCrypt, ASE, FSE, OOPSLA, PLDI, POPL, FOCS, SODA, STOC, UbiComp, UIST, IROS, RSS, EuroSys, USENIX ATC, SIGIR, ISMB, RECOMB, etc. That is, IEEE TVCG/VIS and IEEE TVCG/VR have higher H-5 index than more than 75% of CS conferences listed. They are also better attended than most of the highly selective CS conferences listed as well, with acceptance rates between 12% - 25% (again lower than most of the selective CS conferences listed). Yet they're completed discounted, thereby excluding major publications from more than top 50x R1 institutions within US alone, not to mention worldwide. Similarly Computer Graphics Forum (CGF) is a publication venue for Eurographics conference/symposia. Its H-5 is lower than IEEE TVCG and ACM TOG, but still higher than most of the selective CS conferences listed. Eurographics/CGF for computer graphics is comparable ECCV for computer vision. Between 2000-2016, more than 50 R1 institutions also published in CGF as well -- meeting all criteria listed. If this list is to be 'quantitative', then the list of publication venues must be selected at least consistently within the same field. Within Computer Graphics/Visualization/VR, the top conference publications are published as journal papers for a few years. SIGGRAPH-TOG is one pub, just like IEEE TVCG-VIS/VR is one pub as well. Or, you can list Visualization and VR as separate areas as well, as they're major disciplines within CS, esp. given the number of industry positions within VIS and VR today is no less than those for robotics. The top pub venues with each area would still be IEEE VIS/TVCG and IEEE VR/TVCG. Thanks. |
|
I definitely agree with the last comment. Also the Eurographics/ComputerGraphicsForum pair has the dignity of entering this list. Definitely. |
|
A request statement to add IEEE TVCG, IEEE VIS and IEEE VR to the list of top conference pubs.
ACM TOG, which is a journal, is listed as the only journal with ACM SIGGRAPH as the top conferences in the field of Computer Graphics. ACM SIGGRAPH and SIGGRAPH Asia The relationships between IEEE TVCG, VIS and VR are NO DIFFERENT FROM ACM TOG and SIGGRAPH. There is no argument that IEEE TVCG is often considered on par with ACM TOG; in fact IEEE TVCG is THE PREFERRED venue for top researchers worldwide in Visualization and VR. In particular, VR is becoming hotter than computer graphics today in industry and viewed as the next disruptive technology similar to smart phones. Joint collaboration between ACM and IEEE is already under consideration. If ACM TOG (a journal) is listed, then IEEE TVCG/VIS/VR should not be excluded on the basis that it’s a journal. It is the PRACTICE OF OUR FIELD THAT WE PUBLISH TOP CONFERENCE PROCEEDINGS AS PART OF JOURNAL and vice versa (top journal papers are presented at these conferences). In many ways, our field has pioneered this kind of hybrid publication that is now currently being considered for adoption by ACM (e.g. PACM) and other areas within IEEE CS.
The H-5 index (one high-level measure of impact) of TVCG/VIS/VR is higher than those for APLOS, ISCA, Micro, MobiComp, SIGMETRICS, IJCAI, AAAI, EMNLP, NAACL, CRYPTO, EuroCrypt, ASE, FSE, OOPSLA, PLDI, POPL, FOCS, SODA, STOC, UbiComp, UIST, IROS, RSS, EuroSys, USENIX ATC, SIGIR, ISMB, RECOMB, etc. That is, IEEE TVCG/VIS and IEEE TVCG/VR have higher H-5 index than more than 75% of CS conferences listed in csrankings.org, YET IEEE TVCG, IEEE VIS, and IEEE VR publications are not considered as top CS publications according to your ranking, because TVCG is a journal (well, so is ACM TOG, which is listed). For example, https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_computergraphics Similarly, H-5 for these other conferences in different areas in CS are available for comparison: Real-time Systems: https://scholar.google.com/citations?hl=en&view_op=search_venues&vq=Real-time+Systems Security: https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_computersecuritycryptography Software: https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_softwaresystems Theory: https://scholar.google.com/citations?hl=en&view_op=search_venues&vq=Theory+of+computing HCI: https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_humancomputerinteraction Robotics: https://scholar.google.com/citations?view_op=top_venues&hl=en&vq=eng_robotics SuperComputing: https://scholar.google.com/citations?hl=en&view_op=search_venues&vq=SuperComputing
BOTH IEEE VIS and IEEE VR are also better attended than most of the highly selective CS conferences listed at csrankings.org as well, with acceptance rates between 9% - 25% (again lower than most of the selective CS conferences listed at csrankings.org). Both IEEE VIS AND IEEE VR are A-rated conferences, according to CORE - Conference Portal: http://portal.core.edu.au/conf-ranks/?search=Virtual+reality&by=all&source=CORE2014&sort=atitle&page=1 http://portal.core.edu.au/conf-ranks/?search=Visualization&by=all&source=CORE2014&sort=atitle&page=1 Yet IEEE TVCG-VIS, IEEE TVCG-VR are completed discounted, thereby excluding major conference pubs for thousands of my colleagues in our fields from more than 50 top R1-institutions within US alone, not to mention worldwide.
We appreciate the effort in establishing such metric-based ranking. But, the choices of top conferences seem to be based on subjective opinions and informal survey (no different than US News & Report) in ranking which conference publications would be counted or not. This approach would defeat the claimed rigor of CSRankings. From the pure objective metrics, two most recently added conferences, RTSS and RTAS, have much lower attendance level than IEEE VIS and IEEE VR by comparison, and also higher acceptance rates than IEEE VIS and As a Fellow of ACM and IEEE, who has served on (and chaired some) award committees for ACM SIGGRAPH, IEEE ICRA, IEEE VIS, IEEE VR, Fellow Committees, IEEE Computer Society Board of Governors, CRA-W Board of Directors, and chaired more than 25 ACM and IEEE conferences in computer graphics, animation, robotics, to virtual reality, I can vouch for IEEE TVCG, IEEE VIS, and IEEE VR that the quality of papers in TVCG-VIS-VR are on par with other conferences I We believe that it would do a great disservice to our field in Computer Graphics, Visualization, and VR that our top IEEE CS conference publication venues under vgTC are excluded from the list of top CS conferences. Furthermore, it would run contrary to all the quality and consistency metrics stated as the goal and objective to establish such a CS-ranking list. Thank you for your consideration of this request. |
|
I'd like to point out when you consider that at least 50 top R1 institutes must publish in a conference to be listed, then attendance is implicitly used as a criterion. It's not the practice within CS that we publish in a conference that we do not attend regularly. |
|
I fully agreed with Ming's comments and suggestion. I am very surprised TVCG is not included! I published 30 papers in ACM TOG and SIGGRAPH, and published 21 papers in IEEE TVCG, I believe both two Journals are top in graphics. I served as Associate editor-in-Chief of TVCG, and also served as Paper committee of SIGGRAPH ASIA many times, I think TVCG is a very professional, very well quality control, and very good reputation in community. A very important point is that, due to fast growth of internet and social media, Info vis and Visual Analytics are critical important. As pointed out by John Hopcraft, a Turing Award winner, "The underlying science has shifted to include large random graphs, high dimensional data and dimension reduction, researchers have begun to study social networks with billions of vertices". Some research in Graphics and Visualization is becoming core part in current computer Science. And TVCG is the only top journal for Info Vis and Visual Analytics. |
|
There appears to be a random bug after ACM TOG is taken off the list, with only SIGGRAPH left on the Conference list. However, SIGGRAPH does not publish any regular conference papers since 2002. All regular SIGGRAPH papers are published as journal papers in ACM TOG since 2002. If the code is correct, all computer graphics researchers between 2006 and 2016 should have ZERO CG publications, as there was no SIGRAPH publication for anyone since 2002. But, some still have their TOG papers mostly counted (see attached), while others have lost nearly all their graphics publications, except couple. I'm not sure how any of this is possible if there is no SIGGRAPH paper for ANYONE during this time. There is clearly some strange bug in the system and the code is not counting right.
|
|
Another bug example, my collaborator, Markus Gross at ETH has 40 SIGGRAPH papers counted between 2006-2016 in csrankings.org. According dblp, he has only published 5 SIGGRAPH papers; however, he has published 69 ACM TOG papers.
|
|
Please move the above to a separate issue. TL;DR - CSrankings does not count TOG papers that are not SIGGRAPH proceedings. |
|
Visualization added with e870635 |
IEEE TVCG-VIS/VR probably should be included in the publication list, since IEEE
VIS and IEEE VR are large specialty conferences that each have about 1,000 and 500
attendees annually and their proceedings are published as special issues of IEEE TVCG. There are more than 50 top institutions around the world that have published regularly in IEEE VIS and IEEE VR with IEEE TVCG since 2000.
This arrangement is identical to ACM TOG/SIGGRAPH, which you’ve listed
under Computer Graphics. The CS attendance and participation level at these conferences surpass many system/AI/theory conferences listed under csrankings.org.