Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK) [LWN.net]

Over at Computerworld UK, Glyn Moody compares the (in)famous 1999 Mindcraft study comparing Linux and Windows NT with a recent video that Microsoft has produced. That video shows various folks complaining about OpenOffice.org (and undoubtedly extolling Microsoft Office). "The criticisms made in the video are not really the point - they are mostly about OpenOffice.org not being a 100% clone of Microsoft Office, and compatibility problems with Microsoft's proprietary formats. The key issue is the exactly the same as it was for the Mindcraft benchmarks. You don't compare a rival's product with your own if it is not comparable. And you don't make this kind of attack video unless you are really, really worried about the growing success of a competitor."

to post comments

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 14, 2010 18:30 UTC (Thu) by davide.del.vento (guest, #59196) [Link] (7 responses)

Well,
I think that *Office have the same issues (even to a lesser degree) than their companion presentation tools: they are overbloated with useless features and have the needed ones hard to use. In this sense, yes, OpenOffice (M$ forgot about LibreOffice :-) was really able to catch up with Microsoft :-(
For documents, I still prefer LaTeX!!
For presentations, see here: http://redmonk.com/sogrady/2010/09/14/we-need-a-new-prese...

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 14, 2010 18:36 UTC (Thu) by Trelane (guest, #56877) [Link] (1 responses)

magicpoint?

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 8:11 UTC (Fri) by Np237 (guest, #69585) [Link]

Tried magicpoint once. Oh, the crap. Just as complicated as beamer, but without the functionality.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 19:06 UTC (Fri) by rahvin (guest, #16953) [Link] (4 responses)

I always find it fascinating when someone points out "useless" features. It's been my experience that they taken their very limited world view and usage pattern and apply it to everyone else.

There are a LOT of features in the MS office products that 90% of the world doesn't use, but the 10% that does use them finds them very very useful. A poor analogy would be to point to CADD, although you yourself might find it useless, would you be so quick to point out all the features in it that you wouldn't no how to use as being worthless?

Please consider that there are professions in this world that you know nothing about that have needs that you don't understand.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 19:53 UTC (Fri) by anselm (subscriber, #2796) [Link] (3 responses)

In Office-type programs, one person's useless functionality is another person's indispensable tool. So while the perceived 90:10 split between bloat and essentials is probably true for most users, exactly what is in the 90% bloat and what is in the 10% essentials is very likely to vary from one person to the next.

So even if some altruistic people were to step up and offer to implement the office suite everybody desires, the one which only contains those features that are, in fact, useful, they will probably never reach a consensus as to exactly which features these are.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 23:28 UTC (Fri) by rahvin (guest, #16953) [Link] (2 responses)

That is exactly what I was stating. Thanks for the clear language.

The reality is that we should be seeking to implement the entire feature set of the MS products in FOSS software because MS didn't spend money implementing features no one uses. There are a lot of industries like banking, engineering and accounting out there that needs software to do things you will never ever need if you aren't in those businesses. I don't care for MS or their tactics but honestly they have spent 20 years talking to businesses who use their software and implementing features people need. Sure some GUI stuff and superficial stuff may not be needed or wanted but everyone should be careful saying any feature isn't needed or is bloat. In fact I would argue that the success of office was primarily because they did listen to their customers and implement so many features that the majority of people will never use. If FOSS wants to succeed outside the kid writing a paper for school we need to implement all the features people need even if it's done through a plugin model or other system to reduce bloat.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 17:28 UTC (Mon) by davide.del.vento (guest, #59196) [Link]

I could probably buy the theory that "different people have different needs, and to make everybody happy you do indeed need all those features" that I currently consider bloatware. But for sure the way in which they are implemented today makes them useless to many.
In fact successful software, especially in FLOSS, requires implementing all the features people need (like you said), in a way that they are a joy to use - not copying what MS has done.

Look at Firefox, probably the best example of successful free software: they didn't clone IE at all, reproducing all it (weird) design choices!! Other great examples are the package management systems: in fact, Apple copied that for the iPhone App Store, and something like that is still missing for Windows.
Microsoft "succeed" for 20 years with office, not because they were "listening to customers", but simply because there wasn't an alternative competitive product (same reason they "succeeded" with Windows itself).
Even at the very beginning of its story (as you can read on other comments below)

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 19, 2010 14:54 UTC (Tue) by Wol (subscriber, #4433) [Link]

Actually, the two main reasons behind MS Office's success are (1) it was included cheaply or sort-of-free with new pcs, and (2) most of the competing suites got a bad reputation for reliability - because MS put landmines in the OS :-(

Certainly WordPerfect suffered badly from deliberate sabotage in Windows.

Cheers,
Wol

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 14, 2010 18:45 UTC (Thu) by errare_est (guest, #14275) [Link] (1 responses)

Is this GhandiCon Three?

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 16, 2010 14:17 UTC (Sat) by Baylink (guest, #755) [Link]

Precisely: "Then they fight you."

Presumably to be followed shortly by Four.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 14, 2010 20:09 UTC (Thu) by tialaramex (subscriber, #21167) [Link] (29 responses)

"compatibility problems with Microsoft's proprietary formats"

For Calc at least this isn't a fair characterisation. We have more than adequate documentation for early XLS formats, but Calc can't reliably import complex Excel sheets because the XLS format and the spreadsheet implementation are heavily intertwined (and Microsoft didn't invent this mistake they inherited it from Lotus) such that a deliberately very different implementation like Calc cannot possibly do the Right Thing

For example, XLS (like Lotus) makes no explicit type distinction between a date, a time and a number, and so unless your spreadsheet corresponds exactly to Excel / Lotus in its handling of transformations between representations it will introduce errors that it cannot detect.

And that (believe it or not) is only the beginning. Excel permits sheets to recursively self-execute, (ie you can append results of calculations to a string and then execute that string as part of the formula in a cell) with syntax very different from Calc. There is no way to correctly implement this in Calc without supporting, simultaneously, an entire Excel-compatible formula calculation engine in parallel with its own native engine.

Then you have not one, but two full blown "macro" languages and a powerful extension framework, plus a graphics engine that simply calls directly into GDI, the Windows operating system's low-level graphics system. Before you know it, despite understanding the file format in detail, you are committed to implementing many thousands of man years of unrelated work by Microsoft that was re-used in Excel if you want to load the file.

So, the reason Grandma's shopping list imports fine into OO.o but your 2015 budget forecast that you spend eight weeks creating does not, almost certainly isn't because Microsoft are hiding some secret file format documentation but instead because OO.org (for good reasons or bad) never actually tried to clone Excel and thus can't always do anything useful with the results of correctly parsing the file. A non-proprietary format wouldn't help you.

IMO a clone is a viable project today just as it was in the days of Lotus 1-2-3 (that's what Excel is, a clone of 1-2-3 with a better UI). But evidently OO.o doesn't want to be a clone. Maybe LibreOffice does.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 2:49 UTC (Fri) by bk (guest, #25617) [Link] (24 responses)

The problems are even simpler. Your Excel macros won't work because of basic, seemingly arbitrary differences such as the use of semicolons as argument separators instead of commas, or slightly different arguments to otherwise identical functions (see VLOOKUP).

OO.o is an adequate productivity suite in an alternate universe where MS Office globally disappears in a puff of smoke. Otherwise it will always look like a weird, anachronistic substitute, unless OO.o/LibreOffice teams put in the engineering effort to make it otherwise.

Do they even know what they're running?

Posted Oct 15, 2010 4:19 UTC (Fri) by dmarti (subscriber, #11625) [Link]

I once overheard a conversation between two MSFT Windows users in which one made it pretty clear that she was running OO, but blamed _Microsoft_ for the incompatibility with her professor's documents. I wonder how many people just clicked on the first Google result for [free word processor] and installed it.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 16:43 UTC (Fri) by dbruce (guest, #57948) [Link] (22 responses)

"OO.o is an adequate productivity suite in an alternate universe where MS Office globally disappears in a puff of smoke."

No, it will be "adequate" when people stop thinking "adequate" means "behaves exactly like MS Office".

Around 2000-2003, to be an "adequate" web browser meant "to behave exactly like Internet Explorer". Fortunately, we have gotten beyond the IE-only mentality outside of corporate intranets.

I wouldn't bet on people and companies continuing to pay for MS Office forever. It has enough inertia to go on for years, but once it starts to decline in favor of a free alternative there is no going back.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 23:53 UTC (Fri) by rahvin (guest, #16953) [Link] (21 responses)

I'd argue that we aren't going to get to the point where MS Office leaves the corporate office until those corporations can use the existing spreadsheets they've spent serious money developing.

Whatever the reasons not to, the simple fact is that if LibreOffice follows the path of OO.org and implements their version of Calc as an independent implementation with different syntax than Excel then it's never going to succeed. A lot of software developers simply don't realize how ingrained Excel is in the marketplace including how many people are trained in using it (and don't want to learn new syntax) and how many millions of man-years have been spent developing massive spreadsheets that do the work of dozens of people.

I wouldn't bet on any free replacement having any real success until they are fully compatible with Excel formulas, syntax and the ability to read and use MS Office created .xls (and the new format) files. This includes fully reimplementing every single formula in Excel and all the hundreds of independent plugin's out there that extend the formula library. No Engineer or accountant I know could even think about switching until that happens.

Ordinary people simply don't realize how extensively Excel is used. Let me provide an small example, our local gas provider (Questar Gas) uses a massive (something like 40 sheets) workbook in Excel that they use to predict weather patterns and gas usage so they can plan for the winter heating season. There are literally thousands of formulas ranging from statistical analysis to finite element analysis in this spreadsheet. The workbook takes almost 3 hours to recalculate on a serious computer. They use this spreadsheet every day to make predictions on what people are going to be needing months in advance and work patterns, capital investments and nearly every facet of their operations is impacted by these predictions. Could they run their business if they switched to Calc? Yes, but they would probably need to hire a few hundred people (and raise gas rates significantly) to make those calculations by hand.

Reimplementing the feature set isn't a simple proposition. MS themselves suffer from this problem with every release. Even something like a rather simple GUI change (the ribbon) drew seriously angry comments because it hurt peoples productivity.

I truly believe that if Calc wants to succeed it has to be an Excel clone, at least to gain any significant market-share.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 16, 2010 14:20 UTC (Sat) by Baylink (guest, #755) [Link] (4 responses)

Yup, and this anecdote, far from reflecting poorly on *OOo's implementation of a spreadsheet*, *actually* reflects poorly on that company, for believing that *a spread sheet application* is a good place to implement a complicated (or a database, but that's a different argument) application program.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 17, 2010 17:39 UTC (Sun) by bk (guest, #25617) [Link] (2 responses)

Living in an ivory tower it's easy to say "Well, they shouldn't have done it that way!" yet they have and many more also have workflows that require extensive Excel use. Sorry, that's just how the world is right now.

Either you can bury your head in the sand or (if you care sufficiently) help put in the work to make the open tools better.

These are different problems...

Posted Oct 17, 2010 21:26 UTC (Sun) by khim (subscriber, #9252) [Link] (1 responses)

Living in an ivory tower it's easy to say "Well, they shouldn't have done it that way!" yet they have and many more also have workflows that require extensive Excel use. Sorry, that's just how the world is right now.

Yup. That's unfortunate, but these problems with resolve itself in time like they resolved itself for users of VisiCalc or SuperCalc. Some went with Excel and redid everything, some used saner programs and some are still using decades old programs using some kind of emulator - it's their choice. When/if Microsoft will go bankrupt the same will happen with users of Excel. If people expect that someone else will come and solve problems which they created themselves for free then they don't deserve any sympathy. The people who are hitting real problems doing work with LibreOffice... these are different - there are real imitations and bugs in LibreOffice which can not be easily circumvented. This is where energy and passion must be spent.

Either you can bury your head in the sand or (if you care sufficiently) help put in the work to make the open tools better.

Sure. But why spend time trying to reverse-engineer and implement stuff which even Microsoft can not describe? There are more important problems with Calc.

I'm pretty sure more people are affected by text file disguised as XLS file problem then people who need all these fragile and unduplicable functions and don't have money to pay for MS Office.

We will help these people too if/when Microsoft Office will disappear, but why spend effort on someone who starts talking with you using ultimatum "you must be bug-to-bug compatible with Excel, or else..." ? This is just stupid.

remember when people said that word couldn't replace wordperfect?

Posted Oct 17, 2010 23:51 UTC (Sun) by dlang (guest, #313) [Link]

for some people, word has not implemented all the functionality that they depended on in wordperfect, but that hasn't stopped word from replaceing wordperfect in just about all businesses.

people started using word because of the network effects, that's what the businesses were using, so if they wanted to work with those businesses (or work on that stuff at home) they needed to have word.

even less than 100% compatibility erodes that network effect lock-in

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 20, 2010 5:49 UTC (Wed) by dark (guest, #8483) [Link]

It's working for them. Do you have evidence that they were mistaken in this belief?

Software projects often fail or go over budget. Adding functionality to an existing spreadsheet is a much safer proposition than starting a software project to replace it. And what would the benefits really be?

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 16, 2010 15:04 UTC (Sat) by andrel (guest, #5166) [Link] (11 responses)

This isn't an isolated case. Microsoft have ported Excel to HPC clusters, because there is a large market for it. None of these complicated models will work in OOO.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 17:31 UTC (Mon) by davide.del.vento (guest, #59196) [Link] (10 responses)

This is silly.

If you need HPC power, you need a serious language, with a serious syntax, with several serious testing frameworks, with a serious version control system, ....

Of course Microsoft tries, but if business go that route they'll be doomed to failure.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 19:42 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (9 responses)

Why? A lot of calculations are one-off experiments, to be thrown off shortly after the calculation.

So it makes sense to use Excel (or Python, Perl, whatever). You don't care if it takes 10 minutes instead of 1 second. Because you can whip up an Excel spreadsheet in 20 mins while writing a program in other language make take hours.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 22:02 UTC (Mon) by davide.del.vento (guest, #59196) [Link] (8 responses)

Fine, then you won't need an HPC system, which cost in the millions bucks.

What I'm saying here is that EXCEL on the HPC it's pointless, and here your are confirming what I said: if you don't cares if it takes 10 min instead of 1 second then you don't care to run on a much more expensive (to buy, to maintain, to power, to cool down....) and hard to use (have you ever tried to install scipy on an HPC system? I have, see if you are able to just read the list of what is needed to be done: http://projects.scipy.org/scipy/ticket/1299 ) system.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 22:29 UTC (Mon) by Cyberax (✭ supporter ✭, #52523) [Link] (7 responses)

"Fine, then you won't need an HPC system, which cost in the millions bucks."

First, quite a lot of HPCs are much smaller. And they make a lot of sense if you have a small research institution where a number of people can submit tasks for the HPC cluster. Excel in this case a 'nice to have' feature. Your HPC cluster will, probably, mostly run Fortran code.

And your example about NumPy on AIX actually confirms that MS is right. It's super-easy to manage their HPC clusters (they use familiar ActiveDirectory tools for it). You probably won't be able to approach the raw performance of optimized HPC operating systems, but for a lot of users extra 10% of performance are not essential.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 19, 2010 2:14 UTC (Tue) by davide.del.vento (guest, #59196) [Link] (6 responses)

Even a small system will cost a (significant) fraction of a million, and it will cost more to operate (Windows on it might be easier to manage than AIX, but surely it won't easier than Linux, but that's not the point - are you a M$ saleperson?)

What will you gain with EXCEL on that cluster? Marginal speed increase, if any. For several reasons, including (but not limited to) Amdahl's law. Will be worth doing? Not at all, that money will be best spent learning (or hiring new staff knowing) a serious programming language (e.g. python, if you want faster prototyping) and/or porting your legacy spreadsheets there. My example (about scipy, not numpy!) was there to prove that I do care about a fast prototyping language like python. But it must be a serious language, with a serious syntax, with several serious testing frameworks, with a serious version control system .... not that crappy $-indirection, impossible to test, with very poor version control system that Excel (and its free clones) are.
If you don't understand that, well, discussing with you is moot.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 19, 2010 10:28 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (5 responses)

"Windows on it might be easier to manage than AIX, but surely it won't easier than Linux, but that's not the point - are you a M$ saleperson?"

Actually, it is easier to manage. ActiveDirectory + group policies are very nice. Oh, sure, Linux has more powerful tools (Puppet, for example) but they are not _easier_ for simple tasks.

I hate Microsoft, but dismissing all their products offhand? That's stupidity.

"What will you gain with EXCEL on that cluster? Marginal speed increase, if any."

Quite a large increase, because a lot of people now work on not very powerful notebooks.

"For several reasons, including (but not limited to) Amdahl's law. Will be worth doing? Not at all, that money will be best spent learning (or hiring new staff knowing) a serious programming language (e.g. python, if you want faster prototyping) and/or porting your legacy spreadsheets there."

You don't get it. Excel is _easier_ than 'serious' programming languages for a lot of tasks. It's just faster to visually drop a few data tables onto a worksheet, visually rig up something using Excel's data analysis system and then submit it for calculation. Most of people who use it that way can (and do) program in 'serious' languages.

But I guess you'd spend 10 hours to do a one-off task in a 'serious' language, with complete test coverage and documentation.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 19, 2010 13:48 UTC (Tue) by davide.del.vento (guest, #59196) [Link] (4 responses)

I don't care to discuss ActiveDirectory here that's off topic, so let me go back on topic.

> You don't get it. Excel is _easier_ than 'serious' programming
> languages for a lot of tasks. It's just faster to visually drop
> a few data tables onto a worksheet,
> But I guess you'd spend 10 hours to do a one-off task in a 'serious'
> language, with complete test coverage and documentation.

Ah, ah ah, I can't stop laughing. I can write a python program for the same task of your spreadsheet in half the time. And next week, if we need to change it, I can do it in a quarter time, while you try to figure out what cell contains the code you have to change. Not to mention how difficult it is to share with others: have you ever *tried* to understand somebody's else spreadsheet? With python it's trivial! Have you ever had to use EXCEL on a non-English machine? All the keywords have been translated, often in a silly way impossible to guess even if you are fluent in the other language! I know this for a fact, since my native language is Italian.

The only task at with EXCEL (and its free clones) is faster is when you need very trivial plots, which have a learning curve in python. But as soon as your plots need to be a little more complicated, again python is much faster.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 19, 2010 14:00 UTC (Tue) by Cyberax (✭ supporter ✭, #52523) [Link] (3 responses)

"Ah, ah ah, I can't stop laughing. I can write a python program for the same task of your spreadsheet in half the time."

Wanna bet?

"And next week, if we need to change it"

Most of time you don't need to change it.

"I can do it in a quarter time, while you try to figure out what cell contains the code you have to change. Not to mention how difficult it is to share with others: have you ever *tried* to understand somebody's else spreadsheet? With python it's trivial!"

Have you tried to understand Python script with embedded data tables, spaghetti code and incoherent diagram generation spewing tons of files into a directory with the script?

"Have you ever had to use EXCEL on a non-English machine?"

I speak Russian, Ukrainian, German and Udmurt language and I've used various localized Microsoft products. Is that enough? :)

"All the keywords have been translated, often in a silly way impossible to guess even if you are fluent in the other language! I know this for a fact, since my native language is Italian."

Easy to fix with the 2010 Office. In any case, English keywords are supported even in localized versions.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 23, 2010 1:41 UTC (Sat) by joxn (guest, #70782) [Link] (2 responses)

(Full disclosure: I work for Microsoft. Excel Services for HPC is my product area. And before I came to Microsoft I did fluid dynamics work on Beowulf clusters running Linux, with Python as a webserver interface. I've also used Matlab professionally for image processing. So I have broad experience here, although I'm not exactly unbiased.)

In many cases, there is a substantial backwards compatibility issue which Excel on HPC solves and "rewrite the whole thing in $language_of_the_day" does not.

-- Spreadsheets often are accreted codebases; they started out as a small model in a handy-to-use tool (there are lots of reasons why Excel is handier for this kind of model than Python: visualisation tools, database connectivity, and visual code design are three), with the model expanding as more complex scenarios were envisioned. (Matlab is almost always a better tool for these kinds of toy models. Matlab costs $2000 a seat, while your company has a site license for Excel.)

-- They were often started by people who were domain specialists but not software engineers. Excel formulas and VBA were within their competency set, but Python wasn't; and they didn't have a handy programmer around to ask when they wanted to get a job done.

-- They often have accumulated a lot of business knowledge. Sometimes the business process ends up *defined* by the spreadsheet, not the other way around.

-- Finally, by the time they are large and slow they are often pretty much impossible to rewrite without a major engineering effort. I know of at least one spreadsheet started in 1998 which had grown so big and slow that clearly it needed to be rewritten. There were TWO failed attempts to rewrite it in Java. It is still limping along 12 years later.

Suffice to say, for these reasons there is plenty of market for Excel on HPC Server. I very seriously doubt we're going to lose money on it; the HPC team is very customer focused and doesn't usually build features for which there are no customers. It's actually surprising to me that nobody did it before Microsoft did -- the solution is dead simple conceptually, though of course making it robust and easy-to-use was a challenge.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 26, 2010 11:17 UTC (Tue) by patrick_g (subscriber, #44470) [Link]

>>> Matlab costs $2000 a seat

Why not use Scilab or GNU Octave ?

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 29, 2010 2:27 UTC (Fri) by davide.del.vento (guest, #59196) [Link]

I think departing (see http://money.cnn.com/2010/10/18/technology/microsoft_ray_... ) Ray Ozzie stated in his blog well: Microsoft product's sometimes are too complex and "complexity sucks" http://ozzie.net/docs/dawn-of-a-new-day/ (see the <<Imagining A Post-PC World>> section)

What a load of bull...

Posted Oct 16, 2010 16:15 UTC (Sat) by khim (subscriber, #9252) [Link] (2 responses)

Ordinary people simply don't realize how extensively Excel is used.

Yet ordinary people pay for it. Income from heavy Excel users like your provider is not enough to sustain development of Excel.

I truly believe that if Calc wants to succeed it has to be an Excel clone, at least to gain any significant market-share.

Well, it should stop pretending it's Excel clone when it clearly is not. As for market share... we should fight for new users. Existing users are not that important: most of them can continue to use old versions of Excel indefinitely so if Microsoft will be forced to rely only on sales to existing customers then eventually development of Excel will grind to halt and eventually Excel will disappear.

This means these mystical "highly complicated" spreadsheets which are never distributed are irrelevant. The spreadsheets which ARE relevant are things like price-lists which must be openable by other users - and here we should fight for the switch to open formats like HTML or PDF rather then try to shoehorn them in OpenOffice.org.

What a load of bull...

Posted Oct 17, 2010 3:44 UTC (Sun) by rahvin (guest, #16953) [Link] (1 responses)

If Calc is to be nothing more than for exporting data then why even use Calc? A price list could be distributed a hundred different ways and putting it in a spreadsheet is simply pointless. The whole point of a spreadsheet is formulas and mathematical calculations, otherwise it's just a comma delimited list of data and more appropriate for a database than a spreadsheet.

Spreadsheets are for accounting, engineering and mathematics and the computer version is just a better version of the paper version that's been in use for hundreds of years. They are the method by which we take data and make calculations, that's what makes them a killer application, not data storage. This is what people who don't use spreadsheets for the purpose that made them a killer application don't understand. Calc will never be a success as long as it is just a data storage format with some simple mathematical capabilities. Until it can be used for things like the Questar example above, all it will ever be used for is price lists and other fairly meaningless uses that other formats work equally well for.

Excel became the successor to 1-2-3 simply because MS recognized the engineering, scientific and mathematical capabilities of a spreadsheet and embraced those needs.

Again, wrong question...

Posted Oct 17, 2010 12:33 UTC (Sun) by khim (subscriber, #9252) [Link]

If Calc is to be nothing more than for exporting data then why even use Calc?

For the same reason people are using Excel: it's convenient. 90% of users don't need fancy features of Excel. They use .xls as some kind of "editable PDF". Often these .xls don't contain a single formula - or few very simple ones.

Spreadsheets are for accounting, engineering and mathematics and the computer version is just a better version of the paper version that's been in use for hundreds of years.

Sure, that's why they were created - but only tiny percent of Excel users actually use these capabilities. That's the point.

Calc will never be a success as long as it is just a data storage format with some simple mathematical capabilities.

Calc is already a success (albeit a limited one) and the biggest problem with Calc is not the fact that it's 1-to-1 Excel clone, but problems with UI and layout problems. I cannot print .XLS bill from Calc and be sure it'll look like the bill printed from Excel: in practice it's much worse problem then inability to run complex calculations.

And you can run complex calculations in Calc - using Python and other languages. But it does not mean it needs to be Excel clone: the people who already have Excel-tied spreadsheets surely have Excel license already so why will then want to switch to Calc?

Until it can be used for things like the Questar example above, all it will ever be used for is price lists and other fairly meaningless uses that other formats work equally well for.

And that's quite enough. This is how every disruptive technology replaces it preprocessor. You don't try to make "frontal assault" - this is pointless: it's quite easy for Microsoft to add changes to Excel faster then you can implement them. You start from periphery (from the most unsophisticated users) and make sure group of users which wants to pay for the Excel shrinks over time. Eventually development of Excel will not be profitable anymore, it'll be abandoned and people who are abusing it with three-hour long calculations will be forced to switch to some other product (may be Calc, may be something else).

Excel became the successor to 1-2-3 simply because MS recognized the engineering, scientific and mathematical capabilities of a spreadsheet and embraced those needs.

Sorry, but no. Excel became the successor to 1-2-3 because it was available for Windows first while Lotus spend it's resources first with OS/2 port and then with grand rewrite - and as usual it ended badly. Excel first captured hearts of Apple and Windows users and only when it was market leader it eventually convinced die-hard users of Lotus 1-2-3 to switch. It never offered 100% compatibility (yes, it offered some but more sophisticated packages were not compatible) and it was not a problem.

The problem for Calc is not the fact that it's not 100% Excel compatible. The problem is the fact that it does not work on Android, iPad or ChromeOS. This is where Excel killer will come from - and so far it looks like it'll be something other then Calc...

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 17:41 UTC (Mon) by davide.del.vento (guest, #59196) [Link]

> Reimplementing the feature set isn't a simple proposition.
> MS themselves suffer from this problem with every release.
> Even something like a rather simple GUI change (the ribbon)
> drew seriously angry comments because it hurt peoples
> productivity.

This sounds like: you go to your auto dealer and he says that your car has a problem and need an (expensive) repair. You tell them that you won't do it, because you like it as it is, and don't have money to fix it (forgetting that the price of the car itself, and the price of being towed if it leaves you in the middle of nowhere would be even more expensive).

Is that spreadsheet is mission critical? If not, junk it. If yes, then that business *must* reimplement it in a serious language, with a serious syntax, with several serious testing frameworks, with a serious version control system, .... Will the reimplementation be expensive? Maybe, but it will be more expensive when (note: not "if", but "when") the spreadsheet will break. And while reimplementing it, they can improve their models with better ones, and probably make the model faster (good luck changing *anything* in such a huge spradsheet).

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 16:01 UTC (Fri) by RogerOdle (subscriber, #60791) [Link] (2 responses)

This is exactly my Excel should never be used by anyone seeking long term storage of information. The documents are not ultimately self contained but depend on the ephemeral (i.e. non-standard) features of the operating system. Office should never be used by corporations or governments because the documents do not stand on their own as complete items. OpenOffice is more compatible with Office-95 than Office-2007 is. Why? Because Office is based on document standards? It never was and never will be.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 19:40 UTC (Fri) by bk (guest, #25617) [Link] (1 responses)

This is a bit of an overstatement. Excel formats work fairly well for *storage* of information--you likely will not have difficulty extracting your raw data tables. The problem is with using Excel to *calculate* and/or *analyze* data. These results end up being highly dependent upon the idiosyncracies of the Excel macro engine.

Better to store data in Excel if necessary but do the manipulation and analysis in R. Save the scripts for reproducibility.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 20, 2010 15:44 UTC (Wed) by Trelane (guest, #56877) [Link]

and dependent upon hard/impossible-to-track-down minute changes (e.g. forgot to copy the $ in $A5 into the last three cells in a column, and you then copied the whole range to another location....)

Spreadsheets are nice for some things, but they're a very brittle format for computation with little to no provenance.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 16:54 UTC (Mon) by SpreadWoe (guest, #70689) [Link]

Sorry I'm late to the party...too much RSS to wade through.

Excel issues seem to occur across platforms, too.

I must add that at this very moment one of our Execs is postponing his Win-->Mac OS X migration due to a series of 2007 spreadsheets that do not work correctly on the Mac 2008 version.

Hopefully this...
http://erictric.com/2010/10/17/microsoft-to-make-office-r...
...fixes it.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 14, 2010 20:42 UTC (Thu) by tjc (guest, #137) [Link] (5 responses)

The criticisms made in the video are not really the point - they are mostly about OpenOffice.org not being a 100% clone of Microsoft Office, and compatibility problems with Microsoft's proprietary formats.

The potential for parody is enormous. A company that still supports a product like IE6 shouldn't be antagonizing anyone.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 15, 2010 11:27 UTC (Fri) by wertigon (guest, #42963) [Link] (4 responses)

Microsoft doesn't support IE6 anymore, and in fact has recommended everyone to switch away from it for years now...

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 16, 2010 2:35 UTC (Sat) by rahvin (guest, #16953) [Link] (2 responses)

They say they don't support it but they do actually. There are a LOT of companies out there that have written their entire corporate intra-nets to use only IE6 and to reject any browser other than IE6 (mine is one actually, thank god for user agent switcher in FF).

IE6 is a serious problem in the corporate world actually. I'm sure MS'd like it to drop dead but they are still patching it and still begging everyone to move away from it because it's the source of so many of windows problems. I'm almost betting they are counting more on Windows 7 adoption at the corporate level to kill IE6 than any other efforts they have expended. Last numbers I saw still said IE6 was in the top 3 browsers which is even scarier.

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 16, 2010 11:55 UTC (Sat) by tialaramex (subscriber, #21167) [Link] (1 responses)

Are there really still new IE6 patches? I hadn't seen any indication of that. Do you have a link for something from the last few months (it surely can't have gone that long without /needing/ a patch)?

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 1:19 UTC (Mon) by PaulWay (subscriber, #45600) [Link]

Yes, there are still security patches for IE6. For home users Windows Update practically begs them to upgrade. In corporations, where they can lock down the software versions, they still pay for and get IE6 patches.

The fundamental realisation a lot of FOSS people need to come to is that the corporate market is orders of magnitude larger than the home user market.

And yet FOSS is winning there too. No company I've been in for the last three or four years has dared to try to block FOSS when it's properly recommended and reviewed. The last three companies I've worked for have allowed me to install Linux as my main desktop - in fact it's been vital to doing the job. My current company is looking at offering Linux as a corporate desktop - because they know it saves money. We are winning the battle.

We should give up on expecting Microsoft to capitulate and hand over their business. They're going to go down fighting, like every proprietary software company out there. They'll still lose, it's just not going to be easy.

Have fun,

Paul

Microsoft Gives its Blessing to OpenOffice.org (Computerworld UK)

Posted Oct 18, 2010 10:48 UTC (Mon) by Jonno (guest, #49613) [Link]

IE6 is shipped as a part of Windowns XP and will thus be supported for as long as Windows XP is supported, which will be at least until April 8, 2014.