Talk:Herculaneum papyri

Page move

I don't understand the recent page move by Gluepix. The edit summary says "Change to sentence case (MOS:AT)", but sentence case would mean a lower case "p" - which is what we had before, per WP:NCCPT. Furius (talk) 08:09, 2 August 2023 (UTC)

Completely agree, I've moved the page back, and reverted the WP:GOODFAITH edit by Gluepix. --Yodin^T 10:00, 2 August 2023 (UTC)

#Process comment

§ Process currently describes what the Seales group was up to in 2016. In the context of the Vesuvius Challenge, some parts of the description are quite outdated, even using the Seales group's own updated tutorial for the prize.

• The VC "segmentation & flattening" step is basically the old "segmenting, textualizing, and flattening" in one. What is done is that a curved 3D volume is turned into a straight one aligned with the direction of a patch of papyrus, preserving intensities where possible. It's not trivial, but it just isn't where the new AI stuff comes in.
• The VC has a new "ink detection" step, because it turns out the old method just isn't good enough to produce visually-readable text. The VP people showed in their tutorial (and arXiv 2304.02084) that by feeding multiple layers aligned to the papyrus through a machine-learning model, the model can find a little bit of ink. The tutorial also mentions some (successful, slow) human attempts to find ink. The $40,000 prize is all about this stuff.

Right now we are devoting a lot of text to an outdated way to do S&F. We should be describing what made the ΠΟΡΦΥΡΑϹ word possible: new S&F tooling, a bunch of people doing the S&F, and finally AI engineers trying stuff on the flattened volume. Artoria2e5 🌉 06:30, 13 October 2023 (UTC)

The grand prize for 2023 has now been awarded (I've added it to the article) and the page for it has a lot of detail on how that was achieved, which should be added to the process section. The parts about what the Seales group was doing eight years ago should be reduced to a summary, I think.  — Scott • talk 18:34, 5 February 2024 (UTC)

Bias concerns

Thread retitled from "Possible selfciting by Vito Mocella".

In the "Virtual unrolling"-section, it seems as though Vito Mocella recently has been citing a lot of his own research in a non-neutral tone, and also removed some other material without explanation in the process. Citing own work, in and of itself, is fine according to WP:SELFCITE. However, I have tried to restore a sensible Wikivoice as well as the removed material. (Other changes have been kept. In some places, <ref name="Mocella2015"/> has been tucked onto already cited material, which might be against the spirit of SELFCITE?)

I don't know much about these scrolls, so it might be good if someone checks whether the current version makes sense. This, of course, includes Dr. Mocella himself. I added a notice to the talk page of User talk:VitoMocella68. —St.Nerol (talk, contribs) 21:30, 18 February 2024 (UTC)

Again, I made a partial revert of similar edits. I don't know the facts well, but several things seem biased. For example, the initial section according to which "Several research groups proposed [the method]" was removed (I restored it). Emphasis has been added about the role played by "the team led by Vito Mocella", and that their research "enabled all the subsequent steps". —St.Nerol (talk, contribs) 13:53, 19 February 2024 (UTC)

The reader needs to understand where the innovation lies in comparison to microCT, which has been shown to be incapable of reading papyri. Therefore, a description should be left explaining where the difficulty of microCT lies and why phase contrast is able to overcome these difficulties. when you talk about science, on wikipedia as elsewhere, you use the scientific terminology of major or incremental contribution. If a technique is proven to work, it is a major contribution. If someone else uses it years later, it is an incremental contribution, and of course chronological order counts. The wording you keep coming back is therefore misleading, not to say false say:"Several research groups proposed to unroll the scrolls virtually, using X-ray phase-contrast tomography (XPCT, "phase-contrast CT"), possibly with a synchrotron light source. " and cite as first a work published in late 2016 ! The chronology of events counts and must be clearly established on the page. VitoMocella68 (talk) 14:25, 19 February 2024 (UTC)

I'm reverting, for two reasons. (1) Quality issues. I have until now tried to carefully remove only the most problematic content (biased formulations, botched sentences, strange linking), but even what looks like blatant mistakes keep coming back. (2) More importantly, it looks like you are involved in a feud with Dr. Seales. It seems very inappropriate for you to add things to the effect of "his methods were a failure, mine were a success". And AT THE VERY LEAST, such things should have secondary sources. —St.Nerol (talk, contribs) 20:29, 19 February 2024 (UTC)

Can you explain on what basis you write: "several group proposed... " ? I explained above that you cannot write generically "several groups proposed ..." and put links in a causal chronological order. There is a before and an after, and the Wikipedia page has to respect the order of events in a neutral way. In add, science progresses through unsuccessful attempts! it is not written anywhere this there is a failure and this is a success. There is a link to a B. Seals scientific article (not newspaper!) from 2013 of which the words in it are simply quoted. VitoMocella68 (talk) 16:42, 20 February 2024 (UTC)

Reverting again. Same issues. You could write down here all the actual changes you propose, so that the community can evaluate them one by one.

Also, please do not entirely re-write the heading of this discussion. Would you like to confirm that you are indeed the Vito Mocella whose research is cited in the article? —St.Nerol (talk, contribs) 18:02, 21 February 2024 (UTC)

If you want to know, I am someone who knows and appreciates the work of Vito Mocella and that is why I chose that nickname. Can you tell us who you are instead and what your skills are?? VitoMocella68 (talk) 19:30, 21 February 2024 (UTC)

This is starting to look like an edit war, so I'm pinging a few experienced and recent substantial contributors to the article. @WatkynBassett, NeverBeGameOver, Artem.G, and Scott: Could any of you please give your opinions on this? —St.Nerol (talk, contribs) 18:36, 21 February 2024 (UTC)

I agree, looks like a feud of Vito Mocella and Seales, you might want to escalate it to admins or start RfC. Artem.G (talk) 20:35, 21 February 2024 (UTC)

This is not a feud. This is about chronological order. If I read the (non-scientific) article linked by St.Nerol correctly, it is B. Seals who attacks Vito Mocella, unproven, not the other way round. Instead of making general statements about feud etc., one can answer: how can one write "Several group proposed ... " when the chronology of the scientific articles is clear? Why not answer this simple and clear question ? on the other hand, one only has to read the commentary articles of the 2015 article https://www.nature.com/articles/ncomms6895/metrics or simply The New Yorker paper paying attention to dates:2015 VitoMocella68 (talk) 20:43, 21 February 2024 (UTC)

The topic of this discussion is misleading, the discussion is not about self-citation, but about the correct sequence of experimental papers using phase-contrast tomography, i.e. the first one that showed that the technique worked successfully and the others that confirmed the result. The discussion, if it can be called a discussion since you are not answering on the merits, is about this. Can you answer on the merits? Do you think that citing a paper from late 2016 before one from early 2015, almost two years earlier, and saying "many groups have proposed ..." is just confusing? — Preceding unsigned comment added by VitoMocella68 (talk • contribs) 18:59, 21 February 2024 (UTC)

@VitoMocella68: In my opinion you are engaged in WP:Disruptive editing. I'm considering different venues for WP:Conflict resolution, starting with a notice at Wikipedia talk:WikiProject Classical Greece and Rome#Dispute at Herculaneum papyri. —St.Nerol (talk, contribs) 09:27, 22 February 2024 (UTC)

This is not about a conflict of interest or disruptive edit, but about restoring a clear and unambiguous truth. Does St. have any factual argument to quote, other than an interview by Seals which is clearly not factual? For example, the page in Italian "Papiri di Ercolano", made by someone else supposedly a long time ago, clearly shows the correct primogeniture in the use of the technique. User St.Nerol , on the other hand, claims to quote a generic phrase such as "several group proposed and used phase contrast technique ... " in a chronologically random order and without any factual corroboration. Dear St.Nerol, can you answer with facts ? can you really seriously think that a work published two years later, using the same technique, can be put all together and says "Several group..."  ? VitoMocella68 (talk) 14:48, 22 February 2024 (UTC)

I've now made a notice about the situation at the Conflict of Interest Noticeboard. —St.Nerol (talk, contribs) 14:32, 22 February 2024 (UTC)

For the record, I object to the repeated re-naming of this section by VitoMocella68, most recently to "On the origin of the phase-contrast technique applied to the Herculaneum papyri". This does not describe the contents of this discussion. What I have reacted and objected to is issues of tone, style, bias and self-interest, not factual matters. —St.Nerol (talk, contribs) 16:10, 22 February 2024 (UTC)

St.Nerol, as you yourself said, self quoting is allowed. The issue is exactly what is in the title of the discussion: On the origin of the phase-contrast technique applied to the Herculaneum papyri. You are trying to turn it into a personal matter, without answering it with facts. People reading the wikipedia page do not want to know who wrote this or that, they want to know if what they are reading is correct or not. The question is factual, answer with facts VitoMocella68 (talk) 16:19, 22 February 2024 (UTC)

To be honest, I find it difficult to follow this discussion. For those who work in the humanities, even without being specifically in the field, it is quite undeniable that the 2015 paper, which showed readability using a very sophisticated technique never before used in the field, was a watershed that had echoes in many areas of the humanities. Subsequent developments are certainly very significant and must be acknowledged, but the 2015 work cannot be ignored and its importance must be underlined. Philodemous (talk) 18:32, 22 February 2024 (UTC)

The issue seems to be that we have a RS (this Smithsonian magazine article) that claims the results of Mocella's 2015 paper were greatly exaggerated (i.e. that the letters they claim to have found seem to be noise), and that it doesn't credit Seales' contribution. Now that there's been significant progress by Seales' team, User:VitoMocella68 added Mocella's work to the article in a way that comes across as WP:PUFFERY ("groundbreaking work", "This important discovery, which enabled all the subsequent steps", etc.), and claims the 2015 study revealed letter forms from the scroll, despite this being disputed in the Smithsonian article. When challenged here, their comments come across as quite aggressive (repeatedly claiming that St.Nerol is hiding behind a pseudonym, and may be working for Seales, etc.). I think that the article probably should cover the 2015 paper in a balanced way, but pointed out elsewhere, this should be done by an editor without this clear conflict of interest. --Yodin^T 19:55, 22 February 2024 (UTC)

@Yodin: Yes. Now Philodemous has re-made, with small differences, the same edits as User:VitoMocella68 were doing on the main page (compare Special:Diff/1209613942 with Special:Diff/1209396041). Also the exact same misleading change of title for this discussion (see Special:Diff/1209615732 with Special:Diff/1209566845). This does not seem unbiased at all. Furthermore, Philodemus made his very first edit about three hours after VitoMocella68 was blocked (for username violation, since he denied being Vito Mocella). —St.Nerol (talk, contribs) 21:46, 22 February 2024 (UTC)

I agree: WP:SOCKPUPPET (or WP:MEATPUPPET) seems likely. Probably worth adding to the COI notice, as User:Philodemous will have the same conflict of interests if this is the case. --Yodin^T 23:31, 22 February 2024 (UTC)

I don't have any conflict of interest and I started by the version that appears unbiased. clearly St.Nerol et al. have a huge conflict of interest because reproduce all time the same factually unsupported version and didn't give any explanation and any true answer. However, I add a different pov and some details, for instance on papyrology collaboration in UK, that were probably unknown to previous editors. you should reply with facts. If chronology is incorrect, you should demonstrate, otherwise 2007 appears first but 2015 paper appears before any other phase contrast application! Philodemous (talk) 00:26, 23 February 2024 (UTC)

I feel a certain discomfort in having a dialogue with people who refuse to discuss, who bring no arguments, only unfounded and unjustified personal accusations. The 2015 article is clearly a turning point, it is not an opinion, it is a fact, just look at the echo it has had. Why is that? Because up until then, ink detection had been considered impossible, and every other attempt had been abandoned for several years. The demonstration that it was possible to read instead, albeit with the difficulties then associated with unroll, was a fundamental step. And it was published in a very prestigious scientific journal. Nobody had the arguments to challenge this result, so much so that a few years later another group used the same technique, on the same synchrotron source, and independently confirmed the result Bukreeva I, et al. (2016). There is nothing to balance. There is a hierarchy of source legitimacy. An interview containing an opinion is worth nothing compared to an article in a prestigious scientific journal, unless it is accompanied by published data accepted by independent peer reviewers of the scientific journal. In science, if you want to challenge a result that is more than legitimate, you produce another scientific paper or a commentary in the same journal. Surely the Nature group would welcome any comment if it was scientifically argued. This is not the case; you quotes a simple interview with B. Seals in which he expresses his views. This is an opinion and not scientific result: in science, this interview counts for nothing! B. Seals, as a computer scientist, knows this very well, and if he has not commented, it is because he has not been able to do so. It seems to me to be exactly the same kind of dialogue that you and St. Nerol user in particular) have. You have no argument, not even to counter the evidence of chronological sequence, and then you start with personal attacks. This way of doing things is very suspicious. Philodemous (talk) 10:41, 23 February 2024 (UTC)

References