Talk:Spreadsheet/Archive 2

Neutrality Check
Concerning "neutrality check": the article is currently in a horrible state, much worse than it has ever been before. There's a large number of sections that are, for some reason, about "EditGrid", a very minor "web 2.0" application. Just look at the various subsections in the "Concepts" section: "Real time update", "Refresh rate" are just ads for EditGrid... what do they have to do with generic spreadsheets? It's plain horrible. 131.111.223.43 00:56, 8 June 2007 (UTC)


 * While we are at it, just read -- an amazingly insightful piece... "One person with an agenda can make a big, inappropriate difference on Wikipedia in a seemingly important and widely understood topic that has little participation from those with deep knowledge of the topic" -- goes to the heart of the problem... 131.111.223.43 01:03, 8 June 2007 (UTC)

Neutrality is indeed needed. It is biased towards Web 2.0 offerings.
 * "However, many of these earlier shortcomings can be handled by online spreadsheets such as EditGrid and Google Docs" gives the impression that Google Docs and EditGrid are the future, and other spreadsheet programs are crippled by these shortcomings. Also "Some already surpass them, offering real time updates from remote sources such as stock prices and currency exchange rates" is an ad in disguse. Excel (and others) have offered this functionality since the 90s, and in any case, this feature is inconsequential to this article. While web offerings do have some features that Excel lacks, Excel certainly has many features web offerings lack. Use of the word "already" makes it sounds as if it is only a matter of time before web offerings become superior. This is nothing but speculation. —Preceding unsigned comment added by 89.27.25.177 (talk) 10:04, 20 November 2007 (UTC)

Rewrite, WRS resolution
Despite the last flurry of exchanges, we have seen little in the way of substantive change to this article. It would be nice if somebody would step up to the rewrite project; failing that, do we have consensus that the WRS portion of this article should be removed? Does anybody feel that some of this material should be preserved in a different article, e.g. Imperial Chemical Industries? (If not, should the ICI article be revised accordingly?) Trevor Hanson 19:11, 9 July 2007 (UTC)
 * Remove all associated references. We haven't seen any encyclopedic sources cited for this material. It may be technically interesting, but this type of unsupported content belongs on an external website or blog. Trevor Hanson 19:11, 9 July 2007 (UTC)

Well, I see no further discussion or input here. I guess silence must be taken as assent. I am NOT going to tackle the needed rewrite of this page, but removing the unsourced material seems a necessary step.

I moved some of the WRS content to the ICI page, after substantial redacting. (I attempted to leave material there of a neutral, technical nature, and which seems likely to be accurate; I removed unsubstantiated assertions and inferences about the relationship of that system to the history of computer science. Somebody else can deal with the issue of sources for the ICI page.) I also did some minor editing here, e.g. adding Borland Quattro to the list of important platforms (duh), though not OpenOffice; and I removed a little bloviation.

I encourage somebody else to step in and tackle the remaining historical and technical lacunae. Naturally, if anybody strongly disagrees with what I have done, please step in and discuss it. Trevor Hanson 08:53, 20 July 2007 (UTC)

A few further changes today:
 * Replaced the tag with a rewrite tag, which really seems the right direction at this point.
 * Added an OpenOffice stub section.
 * Flagged the short Spreadsheet Value Rule section as needing restructuring – it's an interesting comment, but like too many other parts of this article, the reader has trouble finding the basic information expected in a "top importance" article on a general computer topic.
 * Update: I realized that this content would fit fine under the 'Values' heading, and moved it. Its original position before the 'Concepts' heading was the problem. At the same time I removed some 'shared public spreadsheet' material, including claims that remote access from spreadsheets is a new invention (it was in use by the early 80s). Trevor Hanson 20:44, 22 July 2007 (UTC)

I'll probably just lurk here now, hoping for somebody else to tackle this. Trevor Hanson 07:52, 21 July 2007 (UTC) Just for the record, I have now scanned in part of the Works records System manual, highlighting some significant points regarding differing terminology.
 * Trevor

http://www.editgrid.com/user/ken/WORKS_RECORDS_SYSTEM

By the way, your reference to "remote access from spreadsheets in use by the 80's" is a little strange since it was used in the "Works Records System" in 1974. Since there were no products I know of that automatically refreshed remote data values into spreadsheets on a network (Other than ICIs own network at least) I find your comment surprising.


 * FYI, my comment was in response to this claim (stress added): "...values can be extracted from a true external source such as the latest stock prices or exchange rates or another public or shared spreadsheet on the same remote server (as in Editgrid). Prior to this very recent innovation, external values could only be obtained from another previously opened and active spreadsheet on the same computer." Surely you'll agree that stock price updates from spreadsheets, for example, is not a recent innovation. I mentioned the 80s simply because I don't have firsthand memory of automatic spreadsheet updates from before then (and there weren't many well-known spreadsheet systems before then either). You may well be correct that WRS was the first such implementation. I was familiar with some other CICS-based WYSIWYG formula-oriented systems from around that time, though I don't recall whether these also dated back to the early 70s. Trevor Hanson 11:18, 14 August 2007 (UTC)

I see that I have been wiped off the face of Wikipedia, despite having been the author of several products that are themselves still on Wikipedia and involved in the production of 3 IMAX movies!


 * I don't think anybody was picking on you. It always came down to the availability of citations from reliable sources on notable topics. (The movie websites, for example, didn't seem to mention you by name.) It's difficult to reconstruct "ancient" computing history with good source material. There's a lot of interesting stuff I personally remember from the 60s-80s that I have no way to post on Wikipedia, because it's all anecdotal. On the other hand, I've observed shouting matches between Wikipedia contributors who essentially were making up their facts, so it's hard to argue with the need for verifiability. BTW I still think it would be a good idea for you to pull all your WRS/ICI history and details together on a blog site, i.e. a site where "original research" and first-person history is appropriate. Perhaps you've already done that though I haven't seen it. Trevor Hanson 11:18, 14 August 2007 (UTC)

Concerning remotely updating stock prices & exchange rates etc, I was referring to "universally available" spreadsheets and built-in functions (i.e. Not VBA, nor add-ons, nor interfaces to external sources) Concerning "Blogs" I did set up something as it happens - which you can see from this link but for some reason it is not to be found by a "Google" search (but then I never did really understand what a blog is and how it differs from the rest of the world!)
 * Trevor

http://worksrecordssystem.blogspot.com/

Apart from "proving" to Dan Bricklin et al, that the WRS really did exist and had very advanced features in 1974 - that can be observed in the few pages I scanned in from the overview documentation (one of three levels of depth) - I have completely lost interest in Wikipedia as an up-to-date source of information. I might as well stick to my dusty old Chambers Encyclopedia - I quite like the smell of old things!

In view of the importance of Spreadsheets to the world and in the name of historical truth, I have suggested to the "museum of computing" that they try to obtain the original source code direct from ICI by appealing to the ICI directors to search their archives for copies of it (it was all held on "Panvalet" libraries which had archiving features). ken 04:02, 15 August 2007 (UTC)

While I haven't completely reread the article, or this talk page; I must say this is a significant improvement over what it was six months ago. 71.218.134.84 (talk) 01:06, 25 January 2008 (UTC)

Best Practice Spreadsheet Modelling Standards
This organization may be perfectly wonderful, but its mention here strikes me like a solicitation for business. Can anybody else weigh in with an opinion on credibility, and even better some independent sources discussing the topic? I've never heard of them, personally.Trevor Hanson 05:48, 28 September 2007 (UTC)
 * Never heard of it, either. I'm finding absolutely no independent reliable sources that support the commentary, nor am I finding any community acceptance that would indicate some sort of notability here.  Since this is essentially a sole source vendor product; with a couple of pretty prominent commercial links to products "if you want to know more", I'm close to just writing this off as simple spam.  Willing to wait a few days to see if I'm just missing something.  Anyone else?  Kuru  talk  12:08, 28 September 2007 (UTC)
 * Still not seeing anything; I've removed the rather terse, unsupported section that is referenced with the promotional material. Again, would love to see more if there is anything.  Kuru  talk  15:29, 6 October 2007 (UTC)

Excel
I suggest that Excel should have a separate expanded section because it is a very popular spreadsheet. --Doopdoop (talk) 23:30, 3 January 2008 (UTC)
 * Since the paragraph includes a link to Excel's article, I believe it to be good enough. However, I think OOo's Calc needs some content. All I see is "its free" and you can get it at Google or its home, where you can get it is completely irrelevant to a spreadsheet article. On an old system (Z-80 CP/M) I used "Report Manager" which was a 3D spreadsheet, and was very disappointed when I upgraded to a 286 and lost those features, modern programs still don't allow you to change view from "Page A" to "Column A" or "Row 1". While the list's usefulness is debatable, if it exists, shouldn't it also have the Mac program Numbers? (I'm not yet ready to BeBold, when I am I'll likely make an account). 71.218.134.84 (talk) 01:06, 25 January 2008 (UTC)

yes its really a great idea.excel should have a seperate expanded section most people works with scenarios, macros, lookup functions, pivot tables, and formulas. Anoopnair2050 (talk) 13:31, 23 July 2008 (UTC)

Data type
I've completely rewritten the 'data type' section. Look, it really was completely wrong. What was referred to is not data typing at all, but was purey data formatting on display. In a computing sense there was no data typing at all (perhaps an argument could be made for one level of dynamic typing). What was described, and what I have rewritten to, is display formatting. peterl (talk) 09:53, 21 March 2008 (UTC)

Excel
The sentence "Excel is now generally considered to have the largest market share." I'm sure that Excel has the largest market share (probably bigger than OpenOffice.org Calc), but seeing some sources would be nice? --BiT (talk) 18:17, 2 February 2009 (UTC)

Cell reference
I understand that R1C1 in Multiplan was innovative (and could be considered more logical, superior to A1), but as VisiCalc and Lotus-1-2-3 were based on A1, the users got used to A1, so A1 won. Excel offers R1C1 but few people use it. I changed "older" in "The older "R1C1" reference style" into "alternative". It is not clear whether R1C1 is actually older and "older" could be wrongly interpreted as if A1 was an improvement from the "older" R1C1.Dominique Meeùs (talk) 22:19, 13 June 2009 (UTC)
 * I'm too young to notice the historical apperance. However, I do see the the fundamental difference between the two conventions. R1C1 is a relative reference, and A1 is an absolute reference. Both are relevant, in certain situations. + m t  04:35, 14 June 2009 (UTC)

Gnumeric "part of the GNOME desktop"?
I don't know what this means. The identical assertion is made within the Gnumeric article and I've commented in its talk page, so I shan't repeat myself here. -- Hoary (talk) 10:43, 31 July 2009 (UTC)

Kanakku deletion
What was the official reason for deleting the reference to Kanakku on this page? Kanakku is a valid web based Spreadsheet application that met the requirements of the page. —Preceding unsigned comment added by Wikiuser009 (talk • contribs) 01:40, 5 October 2009 (UTC)

hi. can you tell the codes that i'd use for making a pivot table? —Preceding unsigned comment added by 196.201.207.51 (talk) 05:16, 28 October 2009 (UTC)

GS-Calc
GS-Calc is a small and fast all-purpose spreadsheet offering 12 mln rows x 4096 columns. GS-Calc supports ODF natively. It can be installed as a fully portable application and run from USB drives. Could you please add the above to the list of spreasheets. Thanks. Jerryp.c5 (talk) 21:45, 24 March 2010 (UTC)
 * I'm sorry, but the spreadsheet must have an actual Wikipedia article before it can be added to this page. Please note, however, that Wikipedia is not for advertising and the program must be notable before it can have an article. TN X Man  21:47, 24 March 2010 (UTC)

Confusing Image Description
The image description for Visicalc says : "Screenshot of VisiCalc, the first spreadsheet.", however there are two other software products mentioned before Visicalc and have the word spreadsheet in their description. Perhaps a better image description would be required ? Or something more clear ? Signing to enable archiving. TN X Man 13:11, 25 March 2010 (UTC)


 * It is not only confusing - it is WRONG!. The first spreadsheet was in fact created in 1974 at ICI Mond Division, in the UK on an IBM mainframe. It was known as "The Works Records System" and had rows, columns and formulae - all designed and entered by non programmers. It ran for 21 years until 2001. Formulae could not get accidntally overtyped (as they can with Visicalc/Lotus 1-2-3 and EXCEL etc) and there were many other superior technical innovations including automatic summations over time periods, database archiving, security, mult-user mode (as standard), The system may well have also been the first to use advanced just-in-time code snippet concatenation and control table techniques etc. The reason none of this is included in the Wikipedia article is Wikipedia's policy of "no original research" (which it seems excludes testimony from all relevant personnel involved - which seems a little extreme and results in many falsehoods being left unchallenged). The original ("status quo") claims are thus disemminated even more widely - through the very medium that seeks to educate the world. The (now archived) edit history of this article covers the systematic deletion of all of the relevant text and links to external sources for this information (see the youtube video "The true history of spreadsheets") for my recent testimonyken (talk) 09:16, 21 May 2010 (UTC)


 * I have moved the rest of the discussion to User talk:Kdakin, since it doesn't directly concern this article. – Adrianwn (talk) 12:36, 23 May 2010 (UTC)

Request for paper spreadsheet image
I have removed the following request from Paper spreadsheets, since it doesn't belong in the article text.

"[An image of a traditional bookkeeping ledger is needed here, exampling how the columns and invoices get listed.]"

Adrianwn (talk) 17:29, 17 June 2010 (UTC)

Clarification and removal of jargon
The article says the acceptance of the IBM PC... began slowly, because most of the programs available for it were "ports from other 8-bit platforms." It would be helpful if this could be translated into mainstream English as the jargon makes it unintelligible to most audiences. Adrian Robson (talk) 16:31, 2 August 2010 (UTC)


 * I wouldn't consider "port" and "platform" jargon, they are the appropriate technical terms; however, internal links are advisable. – Adrian Willenbücher (talk) 06:00, 13 August 2010 (UTC)
 * Well, the whole point of an encyclopedia article is to encoode pefectly plain facts in such a clever way that the reader goes away baffled at the intellectual display set out for his chastisement. A platform is where the train stops, or something you jump from. A port is where a ship stops. An 8088 is also an 8-bit processor, with the bank switching hardware on-chip.  Let's try to write computer articles in English as much as possible; there's so much cant, argot, and jargon in the computer racket anyway, let's not inject marketing-speak until it really is necessary. --Wtshymanski (talk) 14:01, 13 August 2010 (UTC)

A few comments from VisiCalc's co-creator
FYI: I was there as a participant, so I won't edit the main page. I'll just put some comments here hoping others will check it out and make changes if appropriate.

I question the assertion that the IBM PC's acceptance was slow or that it was because of ports. Compared to what? The Apple III? The Apple II? The Radio Shack computers? In big businesses, did you find Commodore 64's? VisiCalc was available for the IBM PC as an "8-bit port" -- you can download a copy of the IBM PC version from bricklin.com where it has been for over 10 years and still runs on some Windows computers. That version is not mentioned in this article. You can read some of the original IBM promotional material about it at http://www.bricklin.com/ibmpcannouncement1981.htm (a longer form of material shown on the Triumph of the Nerds TV program). That version of VisiCalc, available almost immediately after the IBM PC became available in stores, sold very well on the IBM PC until overtaken by Lotus 1-2-3, so there must have been a lot of IBM PCs bought (you wouldn't buy the software just to have it on the shelf). As I recall, the IBM PC version was the largest selling version of VisiCalc, certainly at the time. Where is the backup of the assertion that acceptance of the IBM PC was slow (or a definition of it) or, if it was, because the software was ports of other code? Why would ports of the same program on a better machine not be better than on the original worse machine? Also, Lotus 1-2-3's success is owed to a lot more than just that it was written in 8088 assembly code. There were many very special aspects to the program (both in design and implementation) and its distribution and marketing.

According to the chart on http://arstechnica.com/old/content/2005/12/total-share.ars/4 IBM had dominant market share (the C-64 doesn't count for spreadsheets) in less than a year after first ship. Lotus 1-2-3 didn't ship until after that.

Separately, for the list of current spreadsheets, I guess I should point out that one I wrote in Javascript (no, spreadsheets aren't the only thing I do...) called SocialCalc is included on the One Laptop Per Child's computer, so is probably being used in large quantities, so may want to be mentioned. See http://www.olpcnews.com/software/applications/socialcalc_on_sugar_version_5.html.

Hope this is helpful.

--DanBricklin (talk) 22:39, 16 October 2010 (UTC)

What kind of programming language
I've removed the sentence: "Specifically, they are functional, visual, and multiparadigm languages".

In a sense spreadsheets are not functional, since programs can be constructed that use cells as variables - intermediate states in a semi-automated process. They are not visual: programs are not created by direct manipulation of objects, but by typing formulas. And to assert that they are multiparadigm, one should explain what paradigms it does cover. Diego Moya (talk) 09:25, 8 November 2010 (UTC)


 * I agree on every point, although I would argue that spreadsheets are not functional because they don't support functions as first-class values. But I guess it depends on what definition you use for "functional". Good catch, though. – Adrian Willenbücher (talk) 14:16, 8 November 2010 (UTC)


 * The criteria for being a functional language is that each expression is a call to a function having to side-effects or mutable state, passing parameters that themselves are expressions. Support for functions as first-class values is a feature found in many (but not all) functional languages, but it is not a criteria for being a functional language.  I agree with Diego's critique that extensions to spreadsheets such as the ability in Excel to call imperative VBA code, from which side-effects can be introduced, means that they are no longer pure functional languages.  However, without those extensions -- with just cell-formulas themselves, the claim that they are functional languages is valid IMHO.  I think it is a given that other functional languages are dramatically more expressive/powerful (in a pragmatic sense) and contain many compelling features not found in spreadsheets.


 * There is a visual aspect to spreadsheets -- things are arranged in a visual grid layout. Granted, they are nowhere near as visual as alternative and far superior modeling tools such as Analytica where models are portrayed as distinct visual objects, I think it is clearly the case that the visual layout of information plays a very important role in spreadsheet models. - User:Ldc  —Preceding undated comment added 23:19, 13 November 2010 (UTC).

History of Industry Forms Vs. Spreadsheet History
There seem a one-sided representaion of spreadsheet history !. From the Oil Industry, and generaly, it seems more reasonable that ship's engineer's (or similar) log, or process log sheets, and of course matrix data blocks for early process design assistance, is more likely to `push' the `need' (not +, -, % !). Secondly, if `lazy and non-imaginetive' could assume finace Ind. would name from the product sites `readings' sheets, of flows, tonnage, etc.(just add another column of costs !), and also, if imaginetive would'nt you thick they would come up with a better name, e.g. Cash Ledger (which do exist. No, just a coulpe of columns for finace means don't need to, or can, repeatly open-out sheets, and to `spread' to see. While for navigation estimate, boiler room check, process plant checks, etc, you do, and keep the `records' for obvious reasons, e.g. trending, maintenace, usage, order more feed stock..... — Preceding unsigned comment added by 109.148.255.153 (talk) 14:25, 11 February 2012 (UTC)

Lack of document control as a shortcoming?
In the "shorcomings" section, several references are made to the fact that spreadsheets do not show who made what changes and when, that simultaneous editing by multiple users is problematic, and other issues. I would say that these are not really shortcomings of spreadsheets, in that these are general issues with almost any document saved as a file. Furthermore, Microsoft Excel for example, does have reviewing features that allow changes to be tracked and accepted/rejected. I suppose JPEG files also have a "multi-user collaboration shortcoming" also? — Preceding unsigned comment added by 66.183.6.11 (talk) 18:49, 30 October 2012 (UTC)

Program lists
I took a few cracks at tidying the program lists but there's still total inconsistency in the list and order of program names and/or company names and it makes no sense that, e.g., Gnumeric and Calc are discussed under the "Open Source Software" section while KSpread/Calligra Sheets is under "Other Software". But I'm not conversant enough with all the software to edit all the stuff so if someone wanted to pick up that ball and run with it... --74.162.156.100 (talk) 08:03, 12 May 2013 (UTC)

Spreadsheet risk
Recent event in the media has seen spreadsheet risk gain increasing notoriety and I think it perhaps deserves a more prominent status on the spreadsheet page or a specific page of it own. Thoughts?

DJCS London (talk) 16:47, 6 October 2013 (UTC)

Paper spreadsheets
This article is currently heavily skewed toward recent events; more coverage of the much longer history of paper spreadsheets would be a good addition. -- Beland (talk) 20:23, 26 September 2015 (UTC)

Confusion regarding formulas, conditions and functions
The part of the article dealing with the first IF function example is misleading. The passage begins with the sentence "A formula may contain a condition (or nested conditions)-with or without an actual calculation- and is sometimes used purely to identify and highlight errors." The subsequent example is:

=IF(SUM(A1:A6) > 100, "More than 100%", SUM(A1:A6))

First of all, when SUM(A1:A6) is greater than 1, not 100, the message should be "More than 100%" because percentage values are defined between zero and one inclusively. For example, 0.5 equals 50%, 0.9 equals 90%, 1.0 = 100% and 100 equals 10,000%. This example does not seem to be operating on true percentage values.

More importantly, the code example is not a formula containing a "condition", but rather a formula containing a function, the IF function in particular. And it is the IF function that contains the "condition". It might be true that a formula can contain a "condition", depending on one's definition of the word "condition", but that is not what is shown here.

A formula that contains a "condition" is: =A1 > B1. However, it might be more accurate to say that this formula contains a logical expression.

Baingb (talk) 01:46, 10 October 2015 (UTC)

"Killer App" ?
Why use such a phrase? Spreadsheets were around long before that God-awful phrase was, and quite frankly "must have application" is perfectly adequate and reads far more elloquently and in plainer English. Incidentally, in whose opinion is a the spreadsheet the first "Killer App" anyway? Word Processor's were around before Spreadsheets, AND I would hardly call the early developers and users of them "hobbyists". — Preceding unsigned comment added by MrZoolook (talk • contribs) 12:34, 31 May 2011 (UTC)
 * Ok, lets do this the hard way. Any citation for sales figures of home computers before and after the release of VisiCalc to substantiate the claim that it was more of a must have application then any corresponding word-processor? If not, I will re-edit the article to remove the reference on the basis that the article is not citing a reference to backup that claim in any way. "Mealy mouthed" or not, lets settle for facts instead. MrZoolook (talk) 09:45, 12 July 2011 (UTC)
 * WordStar was the primary reason that many businesses bought the early CP/M systems, and word processors had a much larger user base than spreadsheet programs back in the days of CP/M systems and early PC's. When the IBM AT was released, one popular program was a TSR (terminate and stay resident) app, to remap control, shift, and alt keys so that control function was moved next to the A key the way it was with the earlier PC's, just to make it easier to use WordStar, since WordStar used a lot of control key sequences. Rcgldr (talk) 04:16, 11 November 2015 (UTC)

"Killer App" is the normal terminology within the context of PC software, and it is understood widely to mean a application to drives sales of a platform. Your subjective personal distaste for the term has no relevance to the quality of this article. 69.142.252.103 (talk) 18:24, 1 June 2013 (UTC)

Water is wet. Citation needed.
Spreadsheet users may adjust any stored value and observe the effects on calculated values - this seems to be pretty obvious. Open any major spreadsheet app and check it yourself. Do we really need a citation for that? — Preceding unsigned comment added by 209.63.38.162 (talk) 15:22, 1 June 2016 (UTC)

Works Records System
The document in the archives of the Computer History Museum is not a reliable source. As far as anyone can tell–I have never actually seen this document, and as far as I know you have not either–this is an unpublished document created internally within ICI for its own internal use, and which many decades later one of its former employees has donated to a museum. That is not a reliable source, hence it cannot be used (per WP:V) for this section. Without any reliable sources for the section, the section has to go under WP:V. See also Articles for deletion/Works Records System where this was discussed recently. SJK (talk) 09:54, 27 June 2016 (UTC)
 * Addendum: text claims the system was designed in 1974, cites a 1974 document, then goes on to say that "was in use unchanged for 27 years"–obviously that statement can't be sourced to the 1974 document, since it refers to a period ending 27 years after the document was written. So, even though neither of us have seen the document claimed as a source, it is obvious at least some of the claims of this section aren't even contained in that single unreliable source. SJK (talk) 10:04, 27 June 2016 (UTC)
 * Ah, I see. As much as I hate to see this bit of history getting lost in a blackhole, I must say you have a point there. I spent two hours searching the Internet for Works Records System and things that I found were mostly mirror copies (sometimes, mirror hard copies!) of the deleted Wikipedia article. (I know that because I saw the deleted article on Bing cache.) I wish I lived in Mountain View and could go see that document in CHM. —Best regards, Codename Lisa (talk) 12:07, 27 June 2016 (UTC)