Wikipedia:Articles for deletion/Multilanguage Electronic Publishing System

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was delete. Article's subject is found to not be notable. — Coffee // have a cup // beans // 01:39, 4 March 2015 (UTC)[reply]

Multilanguage Electronic Publishing System

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

the article is about a privately owned publishing system used internally by one publishing company, and is based entirely on primary sources (the only secondary source only makes reference to a separate system), and refers to outdated technical specifications from the 1980s. Jeffro77 (talk) 03:08, 24 February 2015 (UTC)[reply]

Comment—MEPS might be interesting. It might be brilliant, or its function might be greatly exaggerated. As an internal proprietary system, the only sources that say anything about it are produced by the corporation that made it. The article clearly fails Wikipedia's general notability guideline, as the subject has received no coverage in sources independent of the subject.--Jeffro77 (talk) 06:45, 26 February 2015 (UTC)[reply]

Delete for lack of third party sources. Keep – As historically significant. In the 1980s, a system with the ultimate bucky bits keyboard, supporting dozens of languages. There seems to be some information out there. Here is a list of JW documents. Here is a 3rd party mention. Here is a picture of the computer and keyboard. Here is a list of related copyrights. There may be patents too. It was sold by IBM – see Integrated Publishing System, and the cited 1982 Seybold Report. I think a couple of those should be sufficient for the bare minimum keep. I would be interested in more, especially about the software. – Margin1522 (talk) 09:36, 24 February 2015 (UTC)[reply]

Your links to the JW website are primary sources. Of your remaining sources, the 'cloudapp' site mirrors content from the Wikipedia article, and the other only mentions the older system that was sold to IBM. Just to be clear, the Integrated Publishing System that was sold to IBM is not MEPS, it was an older system that was developed separately.--Jeffro77 (talk) 09:43, 24 February 2015 (UTC)[reply]

The link to the JW documents was for anyone interested in finding out what we're talking about, not necessarily for inclusion in the article. I'm not clear myself what rights were transferred to IBM. Probably not hardware. The software, custom fonts, etc.? Those would have been inherited by the MEPS system, so it's not clear to me that they were separate systems. Do you have any other objections, other than being based on primary sources? I don't see any mention of Wikipedia on the the cloudapp page, and the Seybold Report is a secondary source. – Margin1522 (talk) 12:41, 24 February 2015 (UTC)[reply]

It is self-evident that the text at the cloudapp page was copied from the MEPS article; the text relating to MEPS at that page is verbatim from the second paragraph of the History section of the Wikipedia article (in fact the entire page is a collection of extracts of Wikipedia articles containing the word "photo-typesetters"). And as I already stated above, and as explicitly stated at the MEPS article, and in the copy of the text at the 'cloudapp' site, the IPS system sold to IBM and mentioned in the Seybold Report was a different system that was developed separately. So, as already stated, IPS is not MEPS.--Jeffro77 (talk) 13:01, 24 February 2015 (UTC)[reply]

OK. Could you describe how they were different? – Margin1522 (talk) 14:16, 24 February 2015 (UTC)[reply]

The source explicitly states that the independent systems were developed concurrently (however, MEPS was not completed until 1983, whereas IPS had already been sold to IBM by 1982), and that hardware and software was developed separately for MEPS. It is an irrelevant red herring to expect me to know specific elements of the proprietary system beyond those facts already provided. However, one of the primary sources does indeed state the fact that they were developed separately:

To meet these needs, one team of Witnesses began work in Brooklyn on a system that employed a large IBM mainframe computer as well as IBM text entry terminals and an Autologic corporation phototypesetter. Nearly a hundred miles away at Watchtower Farms near Wallkill, New York, the other team started work on an in-house-produced system they called MEPS.—Awake!, 22 April 1984, page 23.

So, as I have already clearly told you several times, and as explicitly indicated above, the systems were developed independently, in entirely separate locations.--Jeffro77 (talk) 14:32, 24 February 2015 (UTC)[reply]

Note: This debate has been included in the list of Technology-related deletion discussions. NORTH AMERICA¹⁰⁰⁰ 14:25, 24 February 2015 (UTC)[reply]

Thanks, but I'm still not certain. The second reference in the IPS article says "IPS, a purely software solution developed by the Watchtower publishing arm of the Jehovah’s Witnesses." It also says "Marketing, licensing, and support for IPS were handled for the Witnesses by IBM, on whose mainframe it was designed to run." So it's unclear whether all rights to IPS were sold. IBM may have been the marketing representative. Also it sounds like IPS was software designed to run on a mainframe, and MEPS was the same software running on JW-built hardware. It just seems rather implausible to me that the Farm team would duplicate the work of designing the fonts and laying out different languages if that work had already been done in Brooklyn. – Margin1522 (talk) 16:50, 24 February 2015 (UTC)[reply]

It doesn't matter what you think "seems rather implausible". What matters is the sources, and they explicitly state that one system was developed on IBM equipment and later sold to IBM, and another system was developed separately and is only used internally. MEPS has no notability outside of the internal proprietary system used by Watch Tower.--Jeffro77 (talk) 02:24, 25 February 2015 (UTC)[reply]

Well, what I think is that this system was historically significant as the most advanced multilanguage typesetting system of its time. It's no accident that it developed by Jehovah's Witnesses, because they are one of the very few organizations interested in creating printed materials for languages with only a few thousand speakers. Another way to look at it is that nobody cares except JW, and therefore not notable. We can disagree about that.

About whether the software was developed independently for MEPS, you keep insisting that it was, but what I see is that separate systems (hardware, software, and peripherals) were developed. I concede that the hardware and peripherals were different. But my view is that since it would take a team of designers several years to develop bit-mapped fonts in multiple pica sizes for 6,000 Chinese and Japanese characters, not mention other scripts, it's unlikely that they did the software twice. In other words, my interpretation of the sources is different from yours. I am not ignoring the sources, I am trying to understand what they say.
All of this would be rather academic, except that it affects whether we get to include the Seybold Report as a third-party source. Neither of us have read it, but this being AfD it's important to determine whether the system attracted outside attention once it became a commercial product. You say no, different system, I say yes, same software. We can disagree about that too, no problem. – Margin1522 (talk) 08:11, 25 February 2015 (UTC)[reply]

Your suggestion that 'only JWs care' is an irrelevant misrepresentation. There is very good coverage on Wikipedia of many topics that are specific to JWs. The fact that the subject 'relates to JWs' is not the issue here. The issue is that MEPS is a proprietary system used internally by a publication company, and it has no notability outside of primary sources. Continuing to mention the Seybold report (September 1982) is also irrelevant misdirection because it says nothing about MEPS, which never became a commercial product. MEPS hadn't even been completed when IPS was sold to IBM in 1982. Awake!, 8 March 1986, page 27 states: "It was not until November 1983 that the first MEPS phototypesetter was put into production use." Also irrelevant is your own unsourced speculation about the development of typefaces, and it is contradictory to imply that the software or the typefaces for IPS were both sold to IBM and also retained by Watch Tower. However, the assertion about typefaces is a red herring, since there is no indication that the typefaces were sold to IBM as part of IPS, and IBM had its own typefaces.--Jeffro77 (talk) 12:06, 25 February 2015 (UTC)[reply]

OK, I will try to explain my view as plainly as I can. I think that the development of the core software for the Jehovah's Witnesses' typesetting system was a single effort that began in the 1970s and has continued up to the present day. In 1982, this software was part of a system called IPS, a "purely software solution" which ran on an IBM mainframe. In 1982, IBM entered into an agreement with JW according to which "Marketing, licensing, and support for IPS were handled for the Witnesses by IBM". The quotes are from the reference cited above. Note that, contrary to the text in the article, it does not say that all rights to the software were "sold" to IBM. On the contrary, it implies that JW retained the right to use this software internally. In 1986, this same software was running on a different hardware platform that had been developed inhouse by JW, on a system called MEPS. But it was the same software.

As evidence that it was the same software, I have pointed to the time and expense of developing typefaces. Here are some references for that (and please don't object that these are primary sources, because I'm offering them for discussion only). Here we read that "Fonts had to be created for each language. This was a major project not only because fonts were not readily available then as now, but because each of the many thousands of fonts for the Japanese language had to be created a pixel at a time." Here we read that JW "were evidently the first people in Myanmar to compose and publish literature using computers...The MEPS system, which used elegant Myanmar characters designed at our branch, sent ripples through the local printing industry." And here is a reference to this book, which describes "Typeface Development". Since you have made so many comments on JW topics you might know this book, or even own it. If so you can check page 597 and satisfy yourself that JW did in fact develop typefaces.

All of this is for the purpose of citing the Seybold Report on the JW typesetting system, which is required for an article to survive on AfD. I'm satisfied that the JW typesetting system existed and that the report referred to it, so I'm going to !vote keep. – Margin1522 (talk) 17:15, 25 February 2015 (UTC)[reply]

I don't care about your speculation because it is unsourced, and it is wrong. I have already quoted a Watch Tower source that explicitly states that MEPS was developed independentally to IPS, at a different location. Development of both systems began independentally in 1979 at different locations. IPS was a software-only system that was finished by 1982, and it was written for use on existing IBM hardware. It was sold to IBM in 1982 and mentioned in the Seybold Report in the same year before MEPS had even been completed. Development of MEPS included the development of hardware, and then software for that hardware. None of the sources even remotely suggest that the same software were used. However, even if it were conceded (it is not) that MEPS used software from IPS, it still would not mean that MEPS is used outside of its proprietary use within Watch Tower, and still wouldn't have any notability outside primary sources. If I write software that incorporates code libraries from Microsoft Office, it doesn't make my new software notable just because Office is popular, and that would remain the case even if I had substantially contributed to code for MS Office. But as already stated, there is no indication from the sources that MEPS incorporates software from IPS. If there were any source for your claim that MEPS uses IPS software, that source could be used in the article for IPS but does not confer notability on MEPS, which remains an in-house non-commercial proprietary system.

Your continued digression into typefaces is irrelevant, because a) it is neither evident nor necessary that the typefaces would be sold with IPS to IBM and b) not all languages were immediately supported when the system was first developed. The (primary) sources you've indicated specifically indicates that the number of supported languages is extensible. In particular, regarding the source about Myanmar, it only indicates that support for that language had been added by 1989, and has no bearing at all on what typefaces might have been available when MEPS was completed in 1983, or when IPS was sold to IBM in 1982.--Jeffro77 (talk) 20:46, 25 February 2015 (UTC)[reply]

Sure, I admit that I'm indulging in a bit of speculation here. This often comes up in my day job. When you have a vague sentence that can be interpreted in two ways, and have to choose one, then you have to do some research and draw on your field-specific knowledge. We have a couple of rather vague sentences in the sources, and two interpretations. I've offered some evidence for my interpretation (primary sources, but better than nothing) and some field-specific knowledge. For example, in those days, users didn't just abandon large software programs. They kept using them, often for decades. That's why COBOL survived for so long. That may be "speculation", but it's better than nothing. Meanwhile, for your interpretation, you've offered nothing except typographical table pounding and assertions that your view is obviously correct and mine is wrong. Hey, you could be right! And you have the advantage that we are at AfD, where the default outcome is delete and only the delete side gets to say "I demand proof!" All I'm saying is that, to me, my interpretation seems more reasonable and has more evidence, so for now I'm going to stick with it.

About the article itself, it's too bad that we don't have enough material to write about this system in more detail. But there are a few things that we can say about it. Someday a professor giving a class somewhere might be able to use it as an example of how the dissemination of religious texts has been important in the history of translation and multilingual publishing. This has been true ever since St. Jerome and the Polyglot Bibles, and it's still true of JW and their 600-language typesetting system. I think that's interesting, so if possible I'd like to keep it. – Margin1522 (talk) 01:54, 26 February 2015 (UTC)[reply]

Seriously? Do you genuinely not understand that there must be sources supporting your claim for it to be valid? You haven't provided any source that supports your 'interpretation', which despite your claim, isn't supported by any evidence. Even the primary sources you've provided do not support your claims. Your characterisation of the AfD process is incorrect but irrelevant; so far only one person (you) has responded—with quite a lot of "typographical table pounding". Without careful consideration of the fact that your entire argument is based on speculation, the article might even be retained, despite the fact that only primary sources discuss MEPS at all, and no sources support your speculation.

Since its first appearance as a copy-and-pasted copyright violation from a primary source (Awake!, 22 April 1984), the article has been granted a great amount of latitude to remain for as long as it has in the hope that it would be improved by secondary sources to indicate notability, but that has not happened, and is unlikely to ever happen, as it is about an internal proprietary system.--Jeffro77 (talk) 06:07, 26 February 2015 (UTC)[reply]

With all of this I've been trying to make a very simple point. The statement "the subject has received no coverage in sources independent of the subject." in the nomination is wrong. It has been covered by the Seybold Report. Whether one third-party source is enough (usually it isn't, but sometimes it is) will be up to the closer and the other !voters. – Margin1522 (talk) 09:46, 26 February 2015 (UTC)[reply]

Please stop repeating this lie. The Seybold Report from September 1982 does not mention MEPS. The Seybold Report was about IBM's use of IPS, which it had acquired from Watch Tower in 1982. MEPS was still in development at the time; MEPS wasn't used in a production environment until November 1983 (Awake!, 8 March 1986, page 27: "It was not until November 1983 that the first MEPS phototypesetter was put into production use."), and wasn't completed until 1986 (Awake!, 22 February 1987, page 27: "In 1979 a team of Jehovah’s Witnesses at Watchtower Farms began developing a computerized system, called MEPS, for producing literature in many languages. By May 1986, when the project was completed"). The JW book Jehovah's Witnesses—Proclaimers of God's Kingdom (which you cited earlier in this discussion) further states that the software also wasn't completed until 1986 (page 596: "By May 1986 not only had the team working on this project designed and built MEPS computers, phototypesetters, and graphics terminals but, more important, they had also developed the software required for processing material for publication in 186 languages.") Your repeated claims that MEPS 'probably used the same software' as IPS does not constitute a reference to MEPS in the 1982 Seybold Report.--Jeffro77 (talk) 09:52, 26 February 2015 (UTC)[reply]

Thank you. That is the first piece of evidence you've provided. So it sounds to me like the team at the Farms began developing their software in the late 1970s, around the same time that Japanese companies like Fujitsu and Hitachi (and later IBM) began developing the first versions of their proprietary software for processing Japanese. (See here, in Japanese) Those companies then deployed their software to a variety of hardware platforms, including mainframes, "office" computers, word processors and PCs. But since I have no proof that JW did the same (although that is by far the most likely scenario) I will have to concede that the Seybold Report could possibly have been about a different language processing system. And since "by far the most likely" is not good enough for AfD, I am changing my !vote to delete. If someone wants to read about this system, they will have to look somewhere else. – Margin1522 (talk) 11:19, 26 February 2015 (UTC)[reply]

Actually, I quoted one of the sources about 24 hours ago that confirmed that MEPS had not even been completed when the Seybold Report wrote about IPS, and the quote about 1986 that I hadn't already provided was from an article you had already cited yourself. Since the sources explicitly state that MEPS was developed by a separate team at a separate location, it is not even 'most likely' that it was the 'same system'. But I appreciate that you are now correctly adhering to what the sources actually say. Thanks.--Jeffro77 (talk) 11:57, 26 February 2015 (UTC)[reply]

The reasons for my !vote are as stated above. – Margin1522 (talk) 15:03, 27 February 2015 (UTC)[reply]

[Redundant response to superfluous comment.]--Jeffro77 (talk) 15:13, 27 February 2015 (UTC)[reply]

Delete. Without independent third party sources it fails the notability test. BlackCab (TALK) 06:54, 26 February 2015 (UTC)[reply]
Delete for lack of appropriate sources (GNG, etc.) as above. — Rhododendrites ^talk \\ 17:05, 1 March 2015 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.