Talk:Manchester Baby/GA1

GA Review

Article (edit | visual edit | history) · Article talk (edit | history) · Watch
I'm proposing to look at this article over the weekend - but not tonight. I see some "polishing" is still underway.Pyrotec (talk) 17:50, 14 November 2008 (UTC)

Initial review

The article is both readable and understandable; and is about the right level for GA.

However, I'm putting the article On hold, for a number of reasons:

1. I wish to ponder the scope of the article, Wikipedia:WikiProject Computing/Early computers task force have a few A-class and B-class articles by no GA-class to provide guidance.
   • I wish to get hold of a copy of Lavington (I should have on on Monday).
   • The picture of the replica shows a "big" computer, so I would have expected some info on factors such as, physical size, power consumption, etc; and of reliability.
2. Lack of precision:
   • In the Background section, the Williams tube is described as a "type of CRT" - I'm not convinced that is precise, it appears to be a standard CRT with an external detector and some external circuitry that provides a capability of "storage" of data.
   • In the Background section a contrast was made with EDVAC; but it is not made clear in that section that development of these computers/computer memory is taking place in parallel. This "confusion" does get explained in the Design and development section, which is the next but one section. So, the Background section needs to be brought "into line", i.e EDVAC and EDSAC is development of computers using existing technology (which has drawbacks) and SSEM is development of new technology for computer use.
   • The Design and development section gives useful details, i.e. four CRTs are used. Yes, but only one is being used as an output display, the other three are configured as Williams tubes.
3. The Background section is short of in line citations in the first paragraph.
4. I think the article needs a bit more tidying up, for example:
   • The flow, or perhaps the section titles, appears to be wrong near the start. We have the WP:Lead; then the Background, the final para of which is about the development of the Williams tube at TRE; then there is a section on the Williams tube, then in the Design and development section we are told that Williams has moved to Manchester and taken some of his old TRE mates along to develop the Williams tube.

These should be resolvable quite quickly.Pyrotec (talk) 12:24, 16 November 2008 (UTC)

Pyrotec says that "The Design and development section gives useful details, i.e. four CRTs are used. Yes, but only one is being used as an output display, the other three are configured as Williams tubes." As I understand it, from converations with a couple of people who were at Manchester at the time that Williams tubes were being used, the distinction between a Williams tube and an output display is false. They were all Williams tubes and could be inspected to see what was going on. Indeed, in one of the machines, there were originally two Willims tubes, one for the accumulator (A) and one for the control register (C). When the idea of an idex register, or "modifier" was introduced, the additional Williams tube was placed between the other two as B and it ws known as the "B-line modifier".TedColes (talk) 12:53, 16 November 2008 (UTC)

You may or may not be right. It would be great if you could provide a verifiable reference. The wikipedia article on Williams tube gives this reference [1] which says that the first detectors were wire mesh and later ones were solid plate so a separate display tube was needed. The paragraph as currently written states that the display CRT acted as the output device and could display the bit pattern of any tube. The only reference given in the paragraph is Lavington (1998), I don't have a copy, and it is not clear whether the reference is intended to apply to the whole paragraph or just the use of EF50 pentodes in the arithetic unit. I'm simply try to verify whether the description is correct and whether the statements are verifiable as per WP:Verify. If you are right, would they not be better described as Williams tubes rather than CRTs?Pyrotec (talk) 15:32, 16 November 2008 (UTC)

I've tried to clarify; all four CRTs were identical, except that the output device didn't have the storage electronics the other three had. Added citation. --Malleus Fatuorum (talk) 22:27, 16 November 2008 (UTC)

The Williams tube did later (post-SSEM) evolve into a type of CRT – some had no phosphor coating on the screen for instance, and different internal arrangements of grids, but the tubes used in the SSEM were standard, commercial CRTs with additional, external electronics. --Malleus Fatuorum (talk) 22:31, 16 November 2008 (UTC)

Replies

• "... type of CRT" rewritten.

• I've moved some material from Design and development to Background, to establish the context of parallel development earlier and put EDVAC into that context.

• I've moved the last paragraph of the Williams tube to Design and development to try and improve the flow.

--Malleus Fatuorum (talk) 14:00, 16 November 2008 (UTC)

Thanks. I'm happy that we can "tick off" these specific holds.Pyrotec (talk) 15:36, 16 November 2008 (UTC)

• I've added some details about the machine's physical size, its power consumption, and the number of valves it used. As it was an experimental machine to test the Williams tube, and quickly evolved into the Manchester Mark 1, I don't think there's any reliability data to be had. --Malleus Fatuorum (talk) 22:34, 16 November 2008 (UTC)

• Added citations to the Background section. --Malleus Fatuorum (talk) 00:30, 17 November 2008 (UTC)

Further expansion

Yes, a big improvement, its almost there on the technical side.

Having just started reading Lavington (1980), it appears that the article misses out important info. It appears that the Ministry of Supply wished to create a National Mathematical Laboratory at NPL. The GPO was too busy to help, so Williams and others from TRE went to visit NPL. Williams was leaving TRE in 6 weeks to take up Professorship at Manchester, so TRE seconded staff to Manchester. It also appears that the Government Chief Scientist initiated the Government contract for Ferranti to make a production version of the Mark I. There also appears to be an NPL/MoS link with ACE, EDSAC and EDVAC. I've also found a bit in IEE Review, May 1998 and IEEE Annals of the History of Computing, Vol 15, No. 3, 1993.

I'd like to try and capture some of this, which will initially mess up the Background and possibly the beginning of the Development and design sections.Pyrotec (talk) 19:47, 18 November 2008 (UTC)

I look forward to that. I agree that the people involved haven't been given sufficient consideration in this article, and that the sociological context would be important for FA. Max Newman, for instance, probably deserves much more credit than he is given here. But this is a GAN, not an FAC. :-) --Malleus Fatuorum (talk) 21:20, 18 November 2008 (UTC)

GAR

GA review – see WP:WIAGA for criteria

1. Is it reasonably well written?

   A. Prose quality:

   B. MoS compliance:

2. Is it factually accurate and verifiable?

   A. References to sources:

   B. Citation of reliable sources where necessary:

   C. No original research:

3. Is it broad in its coverage?

   A. Major aspects:

   B. Focused:

4. Is it neutral?

   Fair representation without bias:

5. Is it stable?

   No edit wars, etc:

6. Does it contain images to illustrate the topic?

   A. Images are copyright tagged, and non-free images have fair use rationales:

   B. Images are provided where possible and appropriate, with suitable captions:

7. Overall:

   Pass or Fail:

Congratulations, I'm awarding GA.Pyrotec (talk) 21:12, 18 November 2008 (UTC)