EDIT: After lots of helpful feedback in this thread, the article is now accepted. Thanks everyone who helped!
Any #Wikipedia editors around who can help? We are trying to get the article on #CHERI added. It's so far been rejected three times:
First, it did not have enough independent citations. We added a lot to news articles about CHERI.
Second, it was insufficiently detailed and lacking context. We added a timeline of development, a load of cross references, and a simple introduction.
It was then rejected again because it lacks an explanation that a 15-year-old could understand. This is true of 90% of science-related articles on Wikipedia, so I'm not sure how we fix it. An explanation at that level is something I can write (I have done for the #CHERIoT book!) but it would then make the page 3-4 times as long and not suitable for an encyclopaedia (I've previously seen pages rejected because Wikipedia is not the right place for tutorials).
I don't understand the standards for Wikipedia and I really need some guidance for how to resolve and progress this.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall we're not involved with Wikipedia but three consecutive rejections are the kind of thing that would make us wonder whether somebody has a grudge against the subject matter
=> More informations about this toot | More toots from ireneista@irenes.space
@david_chisnall they all seem like reasonable complaints individually, just...
=> More informations about this toot | More toots from ireneista@irenes.space
@ireneista To be clear: I think the first one was definitely deserved. None of us have edited Wikipedia for about 20 years and back then the flow was to add a stub and then extend it, because it's much easier to get people to improve articles than contribute new ones, so that's what we tried to do. That seems not to be the flow anymore.
The second one was probably a valid concern as well. We made a lot of improvements after that.
The third one just seems like an editor being obnoxious for no reason. On the other hand, maybe I'm too close to the material and the explanations need to be clearer. But some guidance on that would help, rather than just 'a 15-year-old, who is reading an article on a subject that requires you to have some understanding of how instruction sets work, would not understand this'.
I've had several books, over 150 articles, and a load of academic papers published, so I have some idea of how editing normally works in a variety of contexts. A content editor who wrote the comment from the last review would not remain working at a publisher for very long.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall ah, yeah. well... good, sort of. maybe it's more of a procedural deficiency then, like there's nothing designed to ensure that editing feedback is actionable, or something like that?
=> More informations about this toot | More toots from ireneista@irenes.space
@david_chisnall @ireneista from a quick read, yes, you are too close to the material.
This article is way overcomplicated, without even answering in the lead: What is this, to whom does it matter and where am i (not you) likely to encounter this.
Most of the article is a technical description that is not useful for most readers and can just as well be read from the sources quoted to those who care about it.
=> More informations about this toot | More toots from TheDJ@mastodon.social
@david_chisnall @ireneista And yes this often goes for some of the math articles as well, but they tend to have wider relevance and more historic documentation, giving them slightly more affordances.
=> More informations about this toot | More toots from TheDJ@mastodon.social
@TheDJ @david_chisnall @ireneista As someone who is farmiliar with a software based capability system I would like to add that the article is strangely vague about how this actually works.
I was expecting to see what changes to the different systems are needed and how the different operations on capabilities are implemented. More in the style of the wikipedia entries for TLB or CPU caches.
=> More informations about this toot | More toots from MaxStriebel@mastodon.gamedev.place
@MaxStriebel @TheDJ @ireneista That's partly because it's about the CHERI architecture not a specific microarchitecture. For example, on big systems we store the tag bits in a separate table and have a memory-side (past the final coherence point) hierarchical tag cache and then they flow along with data in cache lines. On embedded systems, we can simply widen SRAM to hold capabilities alongside data. Morello also has a mode where it stores tags in ECC bits in DDR. There are a lot of possible ways of implementing CHERI that make sense on different microarchitectures.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall I think starting with a non-draft stub with adequate refs to establish notability is still the best way. Though my pref is to start a wikidata item even before that.
The drafts process requires a lot of up front work and easily gets stuck if that work results in something that doesn't feel quite right to the reviewer who happens to look. More like a capricious publication process than wiki. /idle commentary
You've done that work though, keep pushing on it, a needed article!
=> More informations about this toot | More toots from mlinksva@mastodon.social
@david_chisnall the 15yo would userstand objection is ridiculous for the reason you state (90% of science articles would fail). But I also think the intro paragraph since that rejection is pretty approachable, so you could resubmit claiming to have addressed the feedback.
=> More informations about this toot | More toots from mlinksva@mastodon.social
@david_chisnall@infosec.exchange We spent many years trying to correct a video game's release date, just because its author gave the wrong one in an interview done 10-20 years after the fact (which is a more reputable source than the game's own documentation, naturally).
We only managed to resolve it because someone who happened to be an experienced Wikipedia editor walked into our community and proposed to clean up the article for us while dealing with the editor politics.
At this point, I honestly encourage specialized communities to build their own wikis, treating Wikipedia as a kind of "jack of all trades, master of none" place.
=> More informations about this toot | More toots from asie@mk.asie.pl
@david_chisnall To be honest, I think you're almost there. Can you give a toy example to explain what's going on toward the start, like it done with Rolle's theorem? https://en.wikipedia.org/wiki/Rolle%27s_theorem
=> More informations about this toot | More toots from dwmalone@mastodon.ie
@dwmalone I've done another pass through the intro, would you mind taking another look?
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall I wonder if something like this could be useful just after the contents? (I guess you have pictures of what capabilities look like, but I glanced at Fig 1 here https://www.computer.org/csdl/magazine/sp/2024/04/10568212/1XXis8UKgUw )
=> More informations about this toot | More toots from dwmalone@mastodon.ie
@dwmalone Good idea. The pictures in slides 21 and 24 from this deck are probably the right shape. I'll try to work out how add pictures to Wikipedia...
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall Yes - they're just the right idea. I think they give a flavour of what is happening that is a good introduction and a non-expert could take away.
Uploading them to wikimedia commons is probably the easiest option - they have an upload wizard to make things easier.
=> More informations about this toot | More toots from dwmalone@mastodon.ie
@dwmalone I've cleaned up the text and it now includes that picture.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall Seems good to me, based on a skim through. Usually, refs come after punctuation on Wikipedia - that's the only thing that stuck out to me, but it's a simple fix.
=> More informations about this toot | More toots from dwmalone@mastodon.ie
@dwmalone Wow. I went and fixed those because literally every other style guide I've read (US and English) says the exact opposite. Any idea why Wikipedia decided to do the stupid thing (treat references as not part of the sentence that they belong to)?
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall Not that this is responsible for your wiki woes, but I find it surprising that you don’t seem to reference IBM AS/400 anywhere. Are the tagged pointers used by that system really so different from capabilities?
=> More informations about this toot | More toots from acsawdey@fosstodon.org
@acsawdey AS/400 does have some similarities, though the B5500 is a more direct inspiration. We cross-reference tagged architectures and I hoped that would send people to all of the relevant systems.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall Well, @HannahClover and @jessamyn are the first that come to mind, I hope they don't mind me tagging them!
Good luck.
=> More informations about this toot | More toots from jandi@mastodon.social
@jandi @david_chisnall @HannahClover Thanks for tagging me, this article seems mostly fine (keeping in mind I only sort of understand it).
I think my only suggestion at this point would be to try to put some of the "notable" aspects of it up top. Like "Oh Microsoft is using it?" "It's a big part of IoT architecture?" Maybe get some specificity or stats in that last sentence in the lede to answer the "Why should I care?" knee-jerk nerd response. Drop that in and ping me back, I can approve it.
=> More informations about this toot | More toots from jessamyn@glammr.us
@jessamyn 🤩
=> More informations about this toot | More toots from jandi@mastodon.social
@jandi @david_chisnall @HannahClover Other suggestions (even though I think it's fine) would be to tighten it up some just to get it across the finish line, maybe have it be more dull and encylopedia, slightly less detailed (save those notes, add them in slowly later) to what is supported by reliable non-CHERI mentions. Again, this is just "how to play the game" advice, not any critique on the article which seems fine.
=> More informations about this toot | More toots from jessamyn@glammr.us
@jessamyn @jandi @david_chisnall @HannahClover another way to improve the intro paragraph would be to include a sentence or two with a bird's eye view of what this is, which a technically-minded layperson could understand. With more [[blue links]] tying it in to larger concepts.
As it stands now, ("...is a computer processor technology designed to improve security. The hardware works by giving...)" I'm still confused about what it is. How does it relate to architectures like RISC or x86_64?
=> More informations about this toot | More toots from semitones@tiny.tilde.website
@jessamyn @jandi @david_chisnall @HannahClover Rather than describe how it works in the intro paragraph, spend some more time explaining what it is, tying into broader concepts people understand. Save the how it works for the body of the article. Imo that would make it more "encyclopedic".
=> More informations about this toot | More toots from semitones@tiny.tilde.website
@semitones @jessamyn @jandi @HannahClover Thanks. I've tried to incorporate all of those suggestions into the new version, which I've now raised for review again.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall @jessamyn @jandi @HannahClover Reads really nicely now! Nice.
=> More informations about this toot | More toots from semitones@tiny.tilde.website
@semitones @david_chisnall @jandi @HannahClover Looks like that all worked out and it's live now? Nice work.
=> More informations about this toot | More toots from jessamyn@glammr.us
@jessamyn @semitones @jandi @HannahClover Yup. Loads of helpful comments and we now have a good initial version. I’ll keep refining it, but at least it’s a real thing now.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall hey @davidgerard can we do something? CHERI is very much worthy of an article.
=> More informations about this toot | More toots from dysfun@treehouse.systems
@david_chisnall maybe @seav has some tips? I’ve had similar trouble getting my own articles for various projects on Wikipedia (lack of “notability” despite being used by millions every month) and just gave up 😅
=> More informations about this toot | More toots from ianthetechie@fosstodon.org
@david_chisnall Maybe @danyork can help ?
=> More informations about this toot | More toots from stucchimax@social.secret-wg.org
@david_chisnall Hi, longtime Wikipedia editor here. I've offered some suggestions on the Talk page of the draft at:
https://en.wikipedia.org/wiki/Draft_talk:Capability_Hardware_Enhanced_RISC_Instructions#c-Dyork-20250123002500-Clarifying_the_intro_section
Looking at the threads, @jessamyn said something similar.
In your current opening, you drop too quickly into HOW it works rather than explaining why someone should care. I think add that part and you'd be good to go.
(thx for the intro, @stucchimax )
=> More informations about this toot | More toots from danyork@mastodon.social
@david_chisnall @jessamyn @stucchimax
I also notice on your User page that you declare a conflict-of-interest with the CHERI article. THANK YOU for doing so - that kind of transparency is critical.
A question, though... is your company that makes CHERI chips mentioned in this Wikipedia article? Is it one of the ones mentioned in the History section?
=> More informations about this toot | More toots from danyork@mastodon.social
@danyork @jessamyn @stucchimax Yes, we are SCI Semiconductor, the ICENI line is ours.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@danyork @jessamyn @stucchimax Thanks, I didn’t have time to do any editing yesterday but there’s been plenty of actionable feedback so I’ll try to do a proper pass today.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall This showed up the Wikipedia Discord server, I've done a small pass to shorten and condense the article and given some pointers on where to go to improve things.
=> More informations about this toot | More toots from Sodium@infosec.exchange
@Sodium Ah, hi! I just left some comments on the talk page. Your edit introduced a lot of factual errors. I'm not sure what the right thing to do is here now.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall Hi Hi, I've responded on the talk page. My edits were based on trying to simplify and create an overview of what was described in the previous iteration of the article. Feel free to let me know where I was wrong and where I could have improved the explanation.
=> More informations about this toot | More toots from Sodium@infosec.exchange
@Sodium Thanks, I've tried in the talk.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@Sodium I've gone through the text and tried to keep the style, while improving accuracy and adding citations.
I've also fixed all of the places where citations were after full stops. I presume Wikipedia style isn't the opposite of every other style guide in this respect?
=> More informations about this toot | More toots from david_chisnall@infosec.exchange
@david_chisnall Wikipedia has the same style as everyone else :) we do citations after punctuation (unless you need to cite a specific things mid sentence). Looks like somebody already cleaned that up and published it tho 🎉🎉
=> More informations about this toot | More toots from Sodium@infosec.exchange
@Sodium I made the change back. No other style guide puts citations after a sentence, they always go with the thing that is citing them. No idea why Wikipedia decided to do this differently, but it’s good to be consistent.
=> More informations about this toot | More toots from david_chisnall@infosec.exchange This content has been proxied by September (3851b).Proxy Information
text/gemini