Talk:Weld quality assurance

Name change notes
Note: In November 2011 this article was moved from "Signature image processing" to    "Weld monitoring, testing and analysis ", and on December 24th, 2011 it was moved/renamed to "Weld quality assurance". Please understand talk items with this in mind. North8000 (talk) 16:48, 23 December 2011 (UTC)

Need
Need to locate the other scientific publications that form the foundation for the technology. Need to assess whether brief mentions of SIP might be made in other welding-related articles. Tony 02:40, 10 September 2007 (UTC)

AFD candidate?
The title looks like one company's specific term for an invention of theirs. I found no indication that the subject term is in use anywhere now, and even back then outside of by that one company. The article is written as promotional copy for that company, and now it appears that the company no longer exists.

I'd say retain it's technical content merged into another article, except there is none. The stuff that appears to be technical content isn't, it's just vague sales talk. North8000 (talk) 12:11, 18 August 2011 (UTC)
 * Err ... there are 11 items in the ref list, many of them in scientific journals. The topic is clearly notable. I think you'll find from reading those that the term is not some company's, but the research engineer's. Tony   (talk)  12:38, 18 August 2011 (UTC)


 * I'll be gone for 4 1/2 days. There's no hurry so might be good to discuss then.   But my main thought is that the article really doesn't cover the topic; it's really just sales talk. North8000 (talk) 17:04, 18 August 2011 (UTC)
 * What aspects/wording is "sales talk"? Tony   (talk)  04:50, 20 August 2011 (UTC)


 * The tone needs work, but "signature processing" is a well-known technique in manufacturing control. Twenty years ago I was building systems for it myself (gearboxes and power presses though, rather than welding). This article seems to be on a company that has developed a strand of this broad technique with application to welding. Now that seems like a good topic to me: it has coverage for basic notability, the technique is of encyclopedic interest and we do cover businesses within WP without everything turning into content-free spam.
 * I'd like to see the following:
 * Editing of the business aspect to remove any concerns over spam. This is the usual copyediting: stripping the weasels and peacocks, not claiming anything that isn't supportable, not listing office addresses, the name of the vice-president for paperclips, or the non-encyclopedic flannel.
 * Expand the context for signature processing, ideally by linking to (maybe writing) a broad article on the topic. When I was doing this, we had recently released PD sourcecode from the US Navy for sonar processing (post Cold War peace dividend) and we used it to listen to car gearboxes through a contact microphone. By listening to gearboxes with known faults (chipped bearings etc.) our system learned the sounds of a fault, then could recognise them in the future. We never knew what these sounds were (the machine did), but the "signature" expressed them in some way that our trained expert system had itself learned to recognise. This is a more useful system than a parametric approach where you predict that if gear #3 breaks, you'll expect a noise of 440Hz, so you listen out for that noise. This is why signature processing works for practical machines, when the parametric approach was too difficult to configure.
 * Describe what they measure, ie. the easily accessible parameter of the process here (which I assume is the electrical behaviour of the welding supply)
 * Describe the welding process, its variables and faults, and which of these are amenable to analysis by their external feature.
 * Describe the algorithmic processing to extract a signature from the measurement, and the categorisation algorithms that can identify behaviours from these varied signatures.
 * Andy Dingley (talk) 10:43, 20 August 2011 (UTC)


 * Some good points, Andy.
 * Where's the weasel? I did go through yesterday and prune obvious offenders.
 * There's no company; I believe it's owned by a university.
 * The article mainly concerns the research.
 * The algorithms are possibly not public: bit detailed even for a WP article. Are you serious about printing them? Not the usual thing, I think. Tony   (talk)  10:53, 20 August 2011 (UTC)

I often write and edit published technical articles. This article is one which I would call absolutely free of technical content, basically a promotional article disguised as a technical article. Rather than me dissecting the whole article, Tony, why don't you pick a paragraph which you feel is actually describing the subject of the article and then I'll chime in on how it isn't?   Sincerely,  North8000 (talk) 02:12, 23 August 2011 (UTC)
 * Refs 4, 9, and 10, plus the patent text, contain the technical details, I suppose. But how technical should the article be? WP's guidelines tend to want articles to be written to a wider readership, provided they're properly referenced. I don't have access to these references any more (not online ... and I don't plan on hiking to the library for them). Do you? Can you point out where the promotional content is? Tony   (talk)  15:36, 23 August 2011 (UTC)
 * My comment would be less about spotting particular items as overly "promotional" and more about the article being nearly 8k long, yet still not communicating what it does. Andy Dingley (talk) 15:40, 23 August 2011 (UTC)
 * I guess it "enables welding processes to be optimised ... using algorithms to assess the quality of the welds produced in robotic manufacturing" bypassing the need for "a definitive understanding of [the complex physical] phenomena [that occur in welding arcs]"—to piece together what the text says. The scientific concept is clear even to me as a non-expert, and greater technical detail is supported in the refs (can you point to other WP articles that print the details of algorithms for such technologies? ... and even if you did find one, does that mean it's necessary to include? The topic is clearly notable on the basis of WP's policy. You keep talking about "the company", but there is none as far as I can see. Tony   (talk)  15:52, 23 August 2011 (UTC)
 * Sorry to be vague: it's quite a few years back, and I picked what looked like a really neat idea, the concept of x-raying welds, so to speak, and building a library of patterns that allow you to identify cracks and other imperfections, from ref 11, which is not a technical journal. I did some leg-work with the other journals at the time, because they were easily at hand, but tried to avoid being what I thought was too technical. But on thinking further about it, the broader concept of signature processing could make a good article: the audio pick-up from gears looks analogous (unsure that it really is, but it could be). Are you guys able to write an article on the broader concept? It looks like one is needed. And although I think this one was the first to use signature images (in more than just welding), I wonder whether sig. imaging has since been applied to other areas now.  Tony   (talk)  16:23, 23 August 2011 (UTC)
 * Tony, I just realized that you wrote the article.  So, first things first, PLEASE PLEASE accept my apology for what is in hindsight the tactless way that I discussed it, and for not checking the history first.  Also, knowing that it was written by someone who was not doing it to promote the company is an important, useful, new piece of information. Sincerely, North8000 (talk) 16:34, 23 August 2011 (UTC)
 * No problem, I think you were right to raise issues; fact is, it needs more referencing and I like the idea of integrating it into a broader topic (if the concept can be made to fit).  Tony   (talk)  17:14, 23 August 2011 (UTC)
 * Thank you so much!   Now on to a few details.  The overall technical field of the is Signal processing.   BTW, it's not Image processing which is processing of multi-dimensional (typically 2) image-type-arrays of information.  More to come. North8000 (talk) 17:48, 23 August 2011 (UTC)
 * Most applications of signal processing to sensing start out and idea or concept which says that a physical parameter (e.g. sound for gearboxes, presumably welding process voltage and/or current waveforms for welding, plus some signal processing method to pull the desired information out of a complex signal.
 * I think that if it is currently being done, that doing this for weld analysis would be / is an excellent article. I'm not knowledgeable in that field; actually I'm not sure that it is a field vs. just one invention. Signature image processing is not a basic description of this...it's sort of a novel/creative way to describe it, this leads to concerns that it may be a term which the company/inventor/owner uses top refer to their particular version.
 * I don't have ready access to any of thos papers.  If someone could email me one of them I'd be happy to read through it and see if I can be helpful in the above areas or if it clears up any questions.    Sincerely, North8000 (talk) 00:18, 24 August 2011 (UTC)


 * Thinking through it now, I'd have though signature processing might end up being a parent article, with signature image, audio signal, etc being offspring articles. Perhaps that's too far into the future. Isn't what astonomers do with light and gravity measurements, and possibly biologists on a nano-scale, signature processing? I don't have enough knowledge to take the lead. Tony   (talk)  01:38, 24 August 2011 (UTC)


 * The "Development" and "Industrial use" sections are OK, for covering what the headers suggest they should cover. However there is still no coverage to explain what this thing is. The refs are not there to explain the function - the article should do this, the refs are there to demonstrate the veracity of this content. If the link exists because the content is only to be found there, not in the article, then they belong under "External links", not "References". The reader should gain their understanding by reading this article, not by just using the article as a reading list.
 * I don't know how this things works. I don't know which welding processes it can be applied to (the answer is not "arc welding" - just read our SMAW or GMAW articles to see what a range there is). I don't know what is being measured directly - voltage / current? I don't know what "image" means here. There's nothing on how a "signature" is generated, on how a signature is evaluated into good or bad, or on how the system is trained to generate such rules. All of these should be answered within this article - that's not asking for proprietary algorithms to be published. Incidentally, what about patents for this, and links to them? As accessible published documents, these are a great resource for WP.
 * I fail to see where "image" fits into this at all. Presumably there isn't any sort of camera sensor imaging the arc. Is it a pseudo-image, where some scatter-ploy of V.I distribution is used to build up a two-dimensional representation, which can then be processed by off-the-shelf algorithms already built to process camera images? Andy Dingley (talk) 02:00, 24 August 2011 (UTC)
 * (Added later) Or considering a v or I vs time plot to be an "image" and a "signature"  (e.g. like an ocilliscope view) and processing that.  North8000 (talk) 02:34, 24 August 2011 (UTC)
 * A plot of dependent variable(s) against an independent variable (i.e. time) is a line graph, not an image. It is one-dimensional data, not two-dimensional. Plotting V against I would give a two-dimensional plot, thus a real image, but either or both of these against time is a simpler situation (because time is independent, ie its progress is unaffected by the welding behaviour). V/t graphs could well be a "signature", but not AFAIC a "signature image". Andy Dingley (talk) 09:24, 26 August 2011 (UTC)
 * A picture of the plot would be an image, but that wasn't my point.  I was more trying to guess at how somebody might have thought up putting "image" into the name.  North8000 (talk) 11:02, 26 August 2011 (UTC)
 * You haven't answered the larger structural issue I just raised. Tony   (talk)  02:12, 24 August 2011 (UTC)


 * In industrial automation, a SET of technologies/methods/areas of expertise/specialized equipment is often the "subject". If the idea of using signal processing to asnalyze welds is in use, I think that THAT should be the article.  One of the main fields drawn upon to do it is Signal processing but, per the previous sentence, that does not make this article a subset of signal processing.  Sincerely, North8000 (talk) 02:41, 24 August 2011 (UTC)


 * I found the inventor's university webpage and emailed him. Amazingly, he got back this morning. He agreed the company thing is secondary, and said the article needs updating, here, add these. So I translated from engineeringese. Can you review it. If a parent article is more appropriate to house this, are you willing to write it? Tony   (talk)  03:29, 26 August 2011 (UTC)


 * what looked like a really neat idea, the concept of x-raying welds, so to speak, and building a library of patterns that allow you to identify cracks and other imperfections, 
 * That sounds like a fine approach to studying welds, and it would clearly fall under "signature images". If you learn to weld (I have basic qualifications for gas, stick & MIG) the split between "amateur" and "apprentice" really begins when formal training requires you not only to weld, but to section and polish your welds afterwards. However that's a destructive testing process - as I read this article, it's a non-destructive in-process quality measure that doesn't require welds to be sampled afterwards. Andy Dingley (talk) 09:27, 26 August 2011 (UTC)
 * The quote—I'm confused about your response to it. I was describing how I came to write the article in the first place. Tony   (talk)  12:46, 26 August 2011 (UTC)

I didn't understand "here, add these. So I translated from engineeringese. Can you review it. If a parent article is more appropriate to house this, are you willing to write it?" and it looks important. Could you explain? Thanks. North8000 (talk) 09:51, 26 August 2011 (UTC)

I also am not sure who wrote "that looked like a really neat idea, the concept of x-raying welds, so to speak, and building a library of patterns that allow you to identify cracks and other imperfections" or what they meant with respect to this article. , and it looks important. Could y'all clarify? Thanks. North8000 (talk) 09:51, 26 August 2011 (UTC)
 * "here, add these."—I was paraphrasing, very quickly, what the inventor said to me in his emailed response. That is, "here are some comments you might add to the article to make it more scientifically explicit, and to bring it up to date."
 * "So I translated from engineeringese."—that is, I translated from his emailed text, which wasn't in WP-type format; I worked out where the expansions would go; I added refs that he gave, using WP syntax and consistent formatting with the pre-existing refs.
 * "Can you review it?" That means, "Can either of you review what I've inserted?"
 * "If a parent article is more appropriate to house this, are you willing to write it?" Means, there was talk above of a broader topic ("Signature processing", I guess): does this still have a bearing on the current article? Tony   (talk)  12:46, 26 August 2011 (UTC)


 * As this article has clearly made it to Wtshymanski's radar, then no, I'm not prepared to waste time on either this or a related article, only to have him follow along afterwards and delete it section by section. Andy Dingley (talk) 14:13, 26 August 2011 (UTC)
 * It seems obvious that it's an important advance in quality assurance for a certain type of welding process. It's supported amply by academic references and other notable publications and broadcasts. Yes, in the larger scheme, a different name might have fitted into the conceptual scheme of things, but I don't see that as an issue that can be dealt with by WP. We just cover what is there. Tony   (talk)  14:17, 26 August 2011 (UTC)

Thanks. Now I understand. Here are some comments and answers to the questions. Those additions are nice. They are also, as would be expected from the owner of the particular technology, very vague  and not informative on the specifics of topic. I don't hold that against him. But it reinforces a key point. What we have here is an article only about one particular person's product, and the title of the article is the name that he gave his particular product. I'm not sure if there are other similar products. I'd be willing and able to absorb / digest any technical material (including highly technical material) that anyone can email me whether it be for the purpose of adding material to the article or to see where we're at. But I'm not ready to take on looking for and obtaining off-line sources. I see the only viable possibilities to be an article on this particular method (as this article is) or on the general topic of weld analysis via signal processing. I think that one of those should exist, which means I don't think we're any longer talking about a potential AFD situation. North8000 (talk) 14:27, 26 August 2011 (UTC)
 * It would be perfectly possible to replace your use of "product" with "science", wouldn't it? The product is just the commercial embodiment of the science, and I believe the scientist/inventor see the science as much more important than that embodiment (and the article name refers—however problematically—to the science, not a product). Apparently the Osaka people have pursued something similar, without acknowledgement. Hmmm .... Tony   (talk)  14:33, 26 August 2011 (UTC)

Unfortunate name
There is no "image", right? No cameras? No 2-d arrangement of pixels whose brightness or color depends on some property of the weld? This is indeed an unfortunate mis-naming of the process, but there's nothing Wikpeida can do about it because it's the original inventor who chose to be confusing. It would be nice to have this clearly explained in the lead. --Wtshymanski (talk) 13:29, 26 August 2011 (UTC)
 * I don't disagree; but even as a layperson, I'd not put such strictures on the definition of "image". They image stars via gravitons nowadays, and it's not an image in the photographic sense. Tony   (talk)  14:19, 26 August 2011 (UTC)
 * If we keep diluting the language like this, eventually we'll all be reduced to pointing and grunting. I'd like "imaging" to mean some process that eventually produces a *picture*, darn it; even if it's using deBroglie waves and tachyon interference in the flux core generators. --Wtshymanski (talk) 14:58, 26 August 2011 (UTC)
 * I guess context matters. In terms of the inherent structure of what is being analyzed, it is one or two variables over time, not a matrix of data (image) One could argue that the graph of the variables vs time is an image, and that is probably what the inventor was thinking when he named his product.  That's his perogotive....he could name it "purple dinosaur" if he wants to.  What has happened is that Wikipedia has taken his chosen trad name and made it the title of the article.  If the article is to be just about his particular method, that's fine.  But if there is other similar weld analysis being done, then it probably would need a broader scope and different title. North8000 (talk) 17:37, 26 August 2011 (UTC)

Article subject, scope and name
I did some technical searches and found a lot. This is basically a large topic which is about 1% covered in Wikipedia, the 1% being the Stephen Simpson "Signature image processing" method covered in this article. North8000 (talk) 17:51, 26 August 2011 (UTC) To open a huge can of worms, here's what I'd consider to be the ideal:

Hypothetical Article: Weld monitoring, testing and analysis


 * 1. Weld Analysis (= after the weld is done)
 * 1.1 Non-destructive weld analysis
 * 1.1.1 Acoustic based
 * 1.1.2 Microscopy based
 * 1.1.3 Imaging Based
 * 1.1.4 Other method
 * 1.2 Destructive weld testing and analysis
 * 1.2.1 Method #1
 * 1.2.2 Method #2
 * 1.2.3 Method #3
 * 2. Weld monitoring (= while the weld is occurring)
 * 2.1 Arc welding electric signal based
 * 2.1.1 Stephen Simpson method, "Signature image processing"   (contents of this article)
 * 2.1.2 Other arc welding electrical analysis method
 * 2.1.3 Other arc welding electrical analysis method
 * 2.2 Visual/imaging-based monitoring
 * 2.2.1 Method #4
 * 2.2.1 Method #5
 * 2.2.1 Method #6
 * 2.3 Other approaches

Seeing the scale of that, and no real expert to write it, my best two suggested realistic ideas are:


 * 1) Leave this article as is, being 2.1.1 from the above. Let someone else worry about the other 99% when they are up to it.
 * 2) Create / morph this into a "Weld monitoring" article ("2." from the above) and leave this article as the only section in it (for now)

Sincerely, North8000 (talk) 18:14, 26 August 2011 (UTC) Finally, if anyone would be able to email me any of the technical papers on this I'd be happy to absorb those and write a "technical core" of this article. North8000 (talk) 18:20, 26 August 2011 (UTC)
 * Sounds like a great advance. I can copy-edit at some stage. Tony   (talk)  01:18, 27 August 2011 (UTC)
 * I'm assuming that you mean one of my final 2 "practical choices"? And if so, which one? Also, are you able to email me any of those papers? North8000 (talk) 13:57, 27 August 2011 (UTC)
 * I'm thinking of doing idea #2. Just to be clear, this will temporarily leave the new article in an awkward position. A bunch of stub sections or sections with headings only (i.e. a framework/skeleton) and then a large "Stephen Simpson method, Signature image processing"  section which would be 90% of this article.  But in the long run I thank that that would lead to creation of the much-needed slightly broader article, and the resolution of the problems with this article, further reinforced by the new tag that someone (correctly) put on.     Any objections? North8000 (talk) 15:00, 12 November 2011 (UTC)
 * I plan to put the following on the talk pages of the persons who posted here in the last 6 months. "Hi, I noticed that you commented at the Signature image processing article talk page.  About a week ago I indicated that I plan/propose to roll this article into a broader article on weld monitoring, testing and analysis, roughly per the outline in my August post. In case haven't been watching the article but would like to give input on this I thought I'd drop you this note.  Sincerely, North8000 (talk) 12:40, 22 November 2011 (UTC) "
 * I did it. So far Tony1 responded, supporting the change/idea. North8000 (talk) 13:54, 22 November 2011 (UTC)
 * Wtshymanski also supported proceeding. North8000 (talk) 18:09, 22 November 2011 (UTC)
 * Not sure on exactly what Andy Dingley meant, but it looks like it's fine with him. North8000 (talk) 21:47, 22 November 2011 (UTC)
 * Andy's been sulking ever since he found out . --Wtshymanski (talk) 22:07, 22 November 2011 (UTC)
 * OK, so here goes! North8000 (talk) 21:48, 22 November 2011 (UTC)


 * Done. I also did some work on it but it needs much much more. North8000 (talk) 10:54, 23 November 2011 (UTC)

At this point then name and scope of the article changed
At this time the name of the article changed from "Signature image processing" to "Weld monitoring, testing and analysis". So all talk prior to this point refers to the previous article. North8000 (talk) 03:00, 15 December 2011 (UTC)

Wikipedia:Articles for deletion/Weld monitoring, testing and analysis
Novaseminary nominated the article for deletion at Articles for deletion/Weld monitoring, testing and analysis Also see article edits/edit summaries for the run-up to this. North8000 (talk) 14:09, 7 December 2011 (UTC)

Found a bit more in depth info on signature image processing
Found a piece on line about signature image processing method. Has much info on the the attributes of the welding process that it utilizes. I left it in just as an external link for now until we can more thoroughly absorb it. North8000 (talk) 12:01, 9 December 2011 (UTC)
 * Novaseminary just deleted it. This article isn't going to be able to develop under such an assault.    I'll put it back in as a source. North8000 (talk) 14:57, 9 December 2011 (UTC)
 * (ec) It doesn't belong in the EL section, so I removed it. But here it is. I'm not sure it is an RS, either, but ”absorb” at will. Novaseminary (talk) 15:00, 9 December 2011 (UTC)
 * May / may not qualify as an RS, but RS is a condition for fulfilling sourcing requirements, not for inclusion of a source. North8000 (talk) 19:51, 9 December 2011 (UTC)
 * I finished absorbing it.  It provides good foundation material for SIP and also briefly intorduces SIP, but does not explain SIP.  A good source fore certain aspects.  North8000 (talk) 15:24, 10 December 2011 (UTC)
 * I support North's handling of this rename/revamp. Nova, please do not make unilateral changes without consensus. Tony   (talk)  13:18, 10 December 2011 (UTC)
 * Last night Novaseminary gutted this article by removing all of the SIP material back into a separate article. I reverted.  North8000 (talk) 13:20, 10 December 2011 (UTC)

I think SIP meets N on its own. If you disagree, send it to AfD. If you disagree with how I summarized it here (with a link to the main SIP article), change it. Novaseminary (talk) 15:37, 10 December 2011 (UTC)
 * No, please stop your disruptive editing, and get a consensus before trying such huge changes. North8000 (talk) 21:28, 10 December 2011 (UTC)

Reliable source
As of this moment, the most cited reference in this article (footnote 2) is this personal website that probably has not been touched in years. This seems to be a copy of part of some version of this Army circular of which there are many versions dating back to at least WWII. At the very least, the citation needs to be corrected, and I would suggest that as is, this link is not an RS. Novaseminary (talk) 06:14, 15 December 2011 (UTC)
 * Nice work finding that new source.  Since the statements that the current source supports are merely that those methods exist, I think that even the current sources is sufficient.  But the upgrade tht you found and suggested is good idea. North8000 (talk) 12:24, 15 December 2011 (UTC)

Possible name change
One hand grenade in the recent barrage had a possible gem in it which, if one puts the article quality first, shouldn't be ignored. Namely, a possible alternate name for the article "Weld quality assurance". I think that that would be a good name change. It's broad enough to encompass the three subjects currently in the title, and a shorter and more elegant title. North8000 (talk) 14:32, 16 December 2011 (UTC)
 * Unnecessary personal attack aside, I'm glad you agree. Novaseminary (talk) 17:14, 16 December 2011 (UTC)
 * Actually it was a complaint about behavior, not a personal attack. But I think one too many in the spirit of new hope. Sorry.  Good source find and good name idea! North8000 (talk) 18:08, 16 December 2011 (UTC)
 * Is there any objection to changing the article name to "Weld quality assurance"? North8000 (talk) 17:47, 17 December 2011 (UTC)
 * Only as far as I think it would be better split, with an article for Weld testing under that name (or some variant, Weld QA qould be fine) and then the bulk of this article returned to an article on the signature analysis technique specifically. We need an article on Weld QA, that's clearly an important topic. Yet the coverage here so far is only at stub level. The signature technique is novel and seems to meet notability on its own, and we already have far better coverage on that. There's enough material for two articles, even at the current level. Andy Dingley (talk) 17:57, 17 December 2011 (UTC)
 * But SIP is just one company's name for for their particular method of using analysis of the voltage and current waveforms and the relationship between them for weld monitoring. North8000 (talk) 18:03, 17 December 2011 (UTC)
 * Yes, and that's enough for an article. Signature processing isn't unique, even for welding, although I've not heard "image" applied to it before outside this article. As it stands, this article is very unbalanced between stub-level content on a huge range, with most of the article focussed on this one novel technique. Andy Dingley (talk) 18:24, 17 December 2011 (UTC)
 * Thaw was the discussed plan, that it would start that way as a way to cause the other material to get built. North8000 (talk) 20:13, 17 December 2011 (UTC)

If SIP meets N, it should get its own article. I think it barely does. I would agree with Andy’s proposal on this. Novaseminary (talk) 21:16, 18 December 2011 (UTC)
 * This is from the person who said that even the broader article which had SIP within it "Further, this seems not much more than a vehicle to promote the academic whose photo appears below and whose work is the only work profiled here". If we were to think about reversing the three month decision process and splitting SIP back off, there would be a lot more to talk about that I'd need more time for.  But it appears that there is no opposition to the rename of THIS article to"Weld quality assurance". North8000 (talk) 12:19, 21 December 2011 (UTC)
 * My main concern was with the promotional nature of the SIP article and the OR-inherent nature of the title (and definition) of this article. By moving this article, and I hope tightening up its lead/definition, we address my concern about this article. We really don't need to discuss the possible SIP article here now. If it meets N, it can be created/recreated/spun-off subject to any disagreeing ed taking that article to AfD (either for a straight deletion, or one of the alternatives for deletion). The question here would be how much of a SIP article should remain here, etc., per WP:SUMMARY. The promotional aspects of the SIP have been tempered somewhat, and could be fixed entirely by editing.Novaseminary (talk) 16:01, 21 December 2011 (UTC)
 * To a technical person, the SIP section has two other problems which are not readily evident. #1 is that the title is sort of deceptive. SIP sounds like a general technical name that their specific technology falls under, but it isn't. While the three word combination may have other random uses, but here it refers to that one particular person/company/organization's methods of weld monitoring. Second, while there is technical sounding stuff in there, it really doesn't explain even the basics on the technology. I planned to write (summarize from the sources) that part if I could get the technicla papers from Tony1, but as it turned out he no longer has them.    So, even the technical sounding stuff is really promotional rather than informative.   Also, as an aside, after a lot of searching, this looks like an idea that that sort of faded out many years ago, although I'm not sure of that.  So, while this probably meets notability and thus could be a separate article, that doesn't necessarily mean that a seperate article is the best choice.   What we decided is that it shouldn't be, but instead become a section in an article on a broader topic which is conspicuously absent from Wikipedia.  I partially condensed the SIP material, but envisioned it being further condensed, but with material possibly added with a true technical explanation if the sources became available. Sincerely, North8000 (talk) 17:39, 21 December 2011 (UTC)
 * I would fully support cutting down, or all but eliminating, the SIP material in this article. (I already tried to do it but you reverted.) Whether SIP gets its own article, though, is based on whether it meets WP:N. Novaseminary (talk) 20:44, 21 December 2011 (UTC)
 * I don't remember that reversion (unless you are referring to the split) but either way I think it should be reduced. I already took out the easy-to-pick stuff. But we should do with a scalpel, not a chainsaw.  :-)  North8000 (talk) 21:32, 21 December 2011 (UTC)
 * Either way I think that we're ready for the name change on this  article. North8000 (talk) 15:25, 22 December 2011 (UTC)

Y Done - moved this article. Novaseminary (talk) 16:34, 23 December 2011 (UTC)
 * Cool. North8000 (talk) 16:50, 23 December 2011 (UTC)
 * Assurance? I'd have though "analysis". I don't see assurance here, not at this earlier part of the pipeline. Tony   (talk)  02:42, 24 December 2011 (UTC)
 * Tony, from a process standpoint, we floated the name idea here for 7 days with no objections before doing. (probably the first thing Novaseminary and I have really agreed on in a year :-) ) It think that it was also responsive to the comments at the AFD against compound titles. But answering your question directly, the thought is that all of those items in the previous title fall under quality assurance, including analysis. Sincerely, North8000 (talk) 11:06, 24 December 2011 (UTC)

Today's substantial additions
The substantial additions made today hav some issues that need to get fixed, but I think that they put/push us a step forward. North8000 (talk) 00:41, 3 December 2012 (UTC)
 * It had more than "some issues" and was entirely unsourced, etc. With this edit I reverted. The material is not "gone", of course. It can be seen and read here. If any of it can be salvaged, cited, and reinserted in an appropriate place rather than stuck in the lead, the world would be a better place. But Wikipedia is better off without it until then. Novaseminary (talk) 05:01, 6 December 2012 (UTC)
 * I didn't like it (as it was) either.  But by "push/put" us a step forward, the area covered by that material is a major area missing from this article, and getting it started, even in that problematic form, was a messy step in the right direction. North8000 (talk) 11:16, 9 December 2012 (UTC)