Slashdot Mirror
Gnumeric Now Supports All Excel Worksheet Functions
unmadindu writes "The latest beta release of Gnumeric has been released. According the the developers, it is now ready and stable enough for general use and deployment, and the final 1.2.0 release will be made on September 8th. This release also marks the realization of a major milestone -- all of the worksheet functions in the U.S. version of MS Excel are now supported. I have been using 1.1.19 for quite some time now, and it is incredibly fast, and hugely improved compared to Gnumeric 1.0."
it could be very cool to support the sxc (openoffice) format. what about this ?
Ploum.net.
Why don't they integrate this thing into OpenOffice?
What about macros, I know that's a major problem with people switching from Microsoft products. Anyone got a VB - php parser? (Or whatever language Gnumeric uses for macros)
Actually meant to say VBA. It's kind of challenging to interpret VBA into a native language, since it's so integrated with Windows. But if a good one was created, then Gnumeric could be fully usable by an Excel user.
--LordKaT
It has many more functions that are not found in Excel, but ALSO having support for everying Excel supports means that any Excel sheet can be opened and used in Gnumeric.
Gnumeric is compatible. It is faster. It does more. That seems better to me, even when ignoring the price tag and lack of Evil(tm) technology.
You can't judge a book by the way it wears its hair.
Microsoft gets a lot of concepts right, they just down implement them as cleanly as most OSS.
So how long before Microsoft chanages Excel to be totally incompatable with their old file format and/or functionality, just to screw the open source community yet again?
It damn well will happen... It's just a matter of how long.
---
Programming is like sex... Make one mistake and support it the rest of your life.
Because few users will switchfrom Excel to Gnumeric if their old files don't work on the new software.
It's like asking why Abiword or Openoffice is spending resources to be able to open .doc files.
...does it support the flight simulator too?
If not, then I'm not interested, thank you!
Oh, and by the way, there's only so many ways to make a usable spreadsheet program. If a standard spreadsheet application exists, and a way of doing things already exists, why reinvent the wheel? This is just so people can be free from the Microsoft grind of upgrading every couple of years to a new, more bloated version of office.
There's still a long way to go though, just because we have all the functions of Excel doesn't mean that it's Excel, or that people won't have functions in Windows (i.e. Macros in VBA) that they need (Like in accounting).
I hope so. The old version that I use my favourite spreadsheet on (a veritable beast, admittedly) slows my box down to a withered crawl.
When the slashdot rush dies down, I'll download it - and see if the dependency frenzy dictates a weeks worth of updates (and a month of fixing once perfect routines), or downloading and installing a new version of linux. And then spend three months to get things working again, before giving up on a couple of them because improvements won't admit them.
Well. It is still better than the alternative. Very.
Well, if it doesn't have all functions of Excel, you can't load all Excel worksheets. So you don't have a full, compatible replacement for Excel. If it didn't have all functions, it would be just a "me too" wannabe.
Lars T.
To the guy who modded me down from perfect to terrible Karma - Apple haters still suck
How long will this be true for? Excel 2003 is coming out soon.
All functions, as defined by their list of functions, is somewhat different than Gnumeric working the same as Excell. For example, I would be amazed if the graphs embedded in spreadsheets and generated from the data look anything like they do in Excell; they certainly were not ever readable in the versions of Gnumeric I've used. Sure, they have a function that calls something that supposedly makes graphs, but the graphs just ain't right. And A.F.A.I.K. this function was on their "already working" list the last time I checked.
I also want to see memos that I've attached to cells in my spreadsheet not vanish when imported into Gnumeric, as well as graphics embedded in a cell. Does anyone know if these now supposedly work?
I'm an American. I love this country and the freedoms that we used to have.
Of course, it doesn't actually support all, or even most, Excel functionality.
It lacks Visual Basic support. And OLE support.
This is really going too help out with Linux deployment - a lot of companies use MS Excel, and now that there is the marketing feature that you can simply copy all your old formula's over and they will work, will be a big deal. I mean Excel is so incredibly bloated now that I bet only 25% of the fomulas are used regually, but at least the marketing dept. can advertise that.
On a side note, how come OpenOffice suddenly became the 'default' linux office suite - Koffice and Gnome Office IMO show much more potential than the overbloated staroffice base...
IntechHosting - Free domain, 2GB, PHP, £4.95/$8.95
...it has all the functions which everybody ignored in Excel.
They say the first thing to go is your penis. Well, it's either that or your brain. I forget which...
So, the U.S. Excel version is missing stuff from other versions? Any idea what?
Worksheet functions are great, but a lot of Excel's draw comes from its embedded VBA. Companies that rely on workbooks with embedded VBA probably wont be willing to switch to Gnumeric until it has support for VBA, or something very similar.
"Open the pod by doors, Hal" > "I'm afraid I can't do that, Dave" sudo "Open the pod bay doors, Hal" > alright
I realize it doesn't have those - but that's for a good reason - those things are freaking huge. OLE and VB would be projects as large as the spreadsheet itself, being made for a fringe group of users (not many people script in Excel - I've done a lot, but few normal users will).
Hell, excel script itself is unpopular because Microsoft includes a warning when you open the file about the script.
Gnumeric Now Supports All Excel Worksheet Functions - followed by this:
all of the worksheet functions in the U.S. version of MS Excel are now supported. Not quite 'All Excel Worksheet Functions' then...
The functions to calculate integrals (need that to calculatr bond rates,) sucked big time in Excell. Insufficient precision.
If you're working on a multi-million dollar, long-term bond that comes to quite a bit of change dropped betwen the cracks.
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
gnumeric is annoyingly slow on my dual 1ghz with 720Mb RAM
There are places where the networks are not touching,and there are places where they are-Boeing's Lori Gunter
Which version of Excel?
i can see a Microsoft ad right below the news :)
Yeah, each of Excel's function names appears somewhere in Gnumeric, but it's far from supporting every formula from Excel. "Supports all Excel worksheet functions" is an extraordinarily silly claim.
+5 Informative for the poor coward, he needs it!
Gnumeric 1.1.20 unleashed!
Date: Friday, August 22 @ 16:36:32 EST
Topic: Gnumeric
This release of GNOME's spreadsheet is _huge_. We've spent the last 2 months squashing bugs, getting the docs in shape, and polishing up the text importer. Trying to summarise the NEWS entries is daunting.
Gnumeric-1.1 is now in beta mode. We'll be doing weekly releases until 1.2.0 comes out hopefully on Sept 8. This includes a string freeze everywhere accept the charting engine. The charts are now somewhat useful, Emmanuel contributed ring, line and area plot support, Jean greased up the gradients, and I tacked on patterns. There are still some rough edges but you can get real work done now.
We also hit a major milestone. All of the worksheet functions in the US version of MS Excel are now supported. The text importer has now been cleanup, and should address all of the common requests for the 1.0.x version.
NOTE : Package this release. It is stable enough for general deployment
and use, for all intents and purposes 1.1.x is now at least as
stable as 1.0 (we hope)
NOTE : http://bugzilla.gnome.org/show_bug.cgi?id=120102
This is a touch worrying. Although the Bonobo build is the default
please drop back to non-bonobo if you run into trouble
Adrian Custer:
* Get the new documentation building.
Andreas:
* Fix analysis tools dialogs again
* Fix text export of non-ASCII characters
* Fix latex export of non-ASCII characters
* Allow latex export to use utf-8 if required
* Avoid duplicate regression tool error messages
* Added new function SSMEDIAN (social science median, or
median for grouped data)
* Added a locale selector to text import.
* Fix 115542 (cs_CZ bug), 115503 (auto-fill of dates)
and many others
* Fix landscape printing
* Save header/footer defaults
* Update consolidate dialog
* Permit to ignore any columns during text import
* Add column header menu to format page of text import
* Fix locale specific parsing on paste
* Add undo to inserting of sheets via the tab context menu
* Clarify insert/append sheet commands
* Change layout of preference dialog
* Make fill->series undo-able, redesign its dialog
* Make Data->Text_to_Columns undo-able
* Add help button to auto-format and plugin manager
* Make Tools->Solver undo-able
* Make Tools->Goal Seek undo-able
* Make Duplicate Sheet undo-able
* Reduce the occasions when the undo chain is being flushed
* Make Edit->Sheet->Remove partly undo-able
* Add standard output selector to consolidate tool
* Make Edit->Fill->Tabulate Dependency undo-able
* Add simple session management
Ben Liblit:
* Another attempt at configure time path generation.
Emmanuel Pacaud:
* New format selector for stf.
* Ring plots
* Line and Area plots
Gustavo Carneiro:
* Use gnome-common detection for libtool version
Jean Brefort:
* Support image files in chart areas (background, bar/col or pie)
* Initial pass at supporting patterns in chart areas.
Jody:
* Fix the generated source files.
* Clean out the internals of PreviewGrid.
* Cleanup Auto-Format dialog.
* Printing support for charting engine.
* Fix printing of images.
* Coup de grace to gal, importing the 75 lines of e_xml we used
* Keynav for auto-format dialog
* Fix jumping to internal links in a different sheet
* Add XML persistence for charts
* Misc Charting improvements & fixes
* Add YEARFRAC to put us at 100% coverage
* Fix some format generation bugs
* Rationalise and document the SheetObject::print in
What about imprecision, then?: good wholesome MS datacorruption...
Other posters ( flat pale people who hang 'round here, i believe ) state that real MS-Excel has significant errors in some operations ( in some uses ). Does gnumeric duplicate that?!?
Wine emulates SM-Windoze bugs, for compatibility, shouldn't we have a globally-broken work-means ( all software ) in order to comply with the Global Standard declared & ordered upon us, by the beeg boiz?
Messages to/for me ( in me journal )
One thing I like about excel is the ability to create complex functions that can be used to get exactly what you want from your spreadsheet data. I know it's possible, but I couldn'tI looked through Gnumeric's online manual to see how it could be extended, but I suppose that part isn't written yet.
How does gnumeric handle excel code that has custom VBA functions integrated into it?
'When the going gets weird, the weird turn pro.' -HST
I've been following gnumeric for a few years now, and I think that it is one of the best OSS projects available currently.
OTOH, OpenOffice spreadsheet works perfectly fine for me for what I use it for
One was Lotus Improv. Radically different than most spreadsheets and far more usable.
The other was called "Advance". I don't remember who put it out, but I think it was inspired by Improv but then went several steps further to produce a tool that was more than just a spreadsheet (thoush you could do most spreadsheet things easily enough) - you could really build complex data models that were very powerful. Indeed, just building the data model was often a very telling and revealing process in itself.
So, why reinvent the wheel?
Because wheels work - and square thingies for the same purpose do not.
And why not reinvent the wheel?
Because in some cases we can do much better.
Or, as Trent Reznor eloquently put it "I've found you can find happiness in slavery".
Calling Bill Gates an innovator (in the grandparent) is like calling Bill Clinton an honorable man.
Get thee glass eyes, and, like a scurvy politician, seem to see things thou dost not.--King Lear
gah???? why would klingon food matter ????
Anyways, I agree with you. I would love to see some common code for doing macros, etc that we could use in the OSS world. If MS wishes to copy it, cool.
I prefer the "u" in honour as it seems to be missing these days.
Just say when and I'll delete my source code and go and buy Excel.
And even on technical merit, Gnumeric is behind in some important aspects, Excel file compatibility the most dire one.
There is a vast difference between
=Sum(A1:A10)
And an Excel.Chart object.
The beauty of Open Source is that, if you feel passionate about these features, you can light off CVS add them, and improve the net happiness of the user community.
Get thee glass eyes, and, like a scurvy politician, seem to see things thou dost not.--King Lear
emerge -p /usr/portage/app-office/gnumeric/gnumeric-1.1.20.e build
I've already got the 2.0beta of Abiword, and it looks great. I've no doubt this will rock too. When you look at those two apps along with Evolution and Mozilla, Linux has GREAT versions of all of the apps that 99% of the population needs/uses.
CB
free ipod and free gmail!
And, please, skip the whining about "that's how floating point numbers work", it's different from Excel, and very significantly different if the result is compared to zero, which isn't that uncommon.
The awful truth is that some very large corporations are run on cobbled together Excel worksheets.
The situation is this: Take some very intelligent people, and provide them with a braindead tool that can, in the end, get the job done. Very few of them will have enough time to find something better, or even to know that there is something better. They will use the tool, and inadvertantly create a nightmare for whoever has to clean up after them.
A multidimensional array of variant, often executable data, with links to a broken-by-design half-object-oriented crudfest of a language, and a horrific hack of the C++ type system, is clearly not the route to computing nirvana.
The world would be a nicer place if these people knew about Python, Haskell, and Prolog, for example, which would accomplish their goals in a cleaner, more efficient and maintainable but ultimately less approachable way.
How do we get this to happen? Education. Only when computing (not "How to use some applications"), and multiple models of computing (procedural/OO, functional, and logical), are taught in schools at a young age ( 11 upwards), as a basic subject as fundamental as other sciences and humanities, will people do things "right" from the beginning.
Will it happen? Doubtful. All we can hope for is that someone comes up with something that strikes a balance, and lets people do their work easily, without creating a horrific mess. Also doubtful.
What happens if some of these functions don't quite work identically to Excel's in 0.1% of cases, be it for a bug in Excel or Gnumeric? I don't see much rush for converting existing work to Gnumeric, just because of this risk factor.
Linux started with GNU gcc version 1.37. Wow, does that seem like a long time ago. There was not even a working curses library at the beginning. Only stuff which relied on the standard C libary could be made to work, and not even all of that.
So while this Gnumeric milestone deserves a "hats off" to all the wizards on the Gnumeric team, let's not forget all those who over the years toiled away at improving the GNU toolchain -- compilers, linkers, libraries, debuggers, and all those who worked to make XFree86 as stable as it is today. They layed the groundwork for Gnumeric and all the great software to come.
From the graphing functions to its statistical capabilities, I consider Gnumeric to be on part with Gimp itself as an example of the quality that the Open Source model can create.
Any idea whether there is a windows version? Now that would be a good idea. I don't know why there isn't more work Open Source development being done for windows. How about giving Microsoft their own taste of "embrace and extend" by using Open Source on Windows as a means of reaching those who aren't likely or able to move over to Linux? I for one was VERY glad to see that Gimp had been ported to windows. I kept getting asked by windows users if there was a good alternative to Photoshop and now I can finally say yes without qualifying my answer with "but it only runs on Unix."
Microsoft isn't nearly as afraid of Linux as it is of the Open Source / Free Software movement/model itself. The technical quality of Microsoft's products is often lackluster, but when it comes to business strategy its leaders are grand-masters. They'll bankrupt you using an inferior product nine times out of ten. So far open source products like Linux have frustrated their ambitions to move up into the enterprise server arena but that isn't the same as going after them in their own backyard. Linux CAN be every bit as useful as a desktop OS as anything Microsoft or Apple has to offer, but it isn't quite there yet. Soccer moms and secretaries simply aren't going to move over to Linux because it isn't what their computers ship with and it isn't what everyone else is using. It also requires a degree of technical acumen that almost no-one posesses. The same is true of Windows of course, but that doesn't work against it since it's already in the dominant position. Those of use who do posess skill and talent with computers often forget just how mysterious the things that seem obvious to us are to most people. That is why Linux is stuck in the server room and will be for the forseeable future. If we can't displace Windows on the desktop, why not use it against its masters? Imagine if all the open-source application work that has been done for Unix was targeted at windows as well? Everyone who owned a computer would be using open source software in some capacity, and many would be aware of it. This would make it much easier to move people off of windows onto something better.
Before this movement to something better can occur however Linux needs to be made more luser friendly. Before you can sell something to someone you have to show how it is better than what they are already using and how what they are using is detrimental to them in some way that the replacement is not. Just making a better mousetrap isn't good enough when your potential customers have already invested in another model. Your mousetrap has to kill more mice AND include a feature whereby human fingers will never be smashed by it accidentally. Right now Linux is comparable to Windows as a desktop os in most ways. It needs to be better than windows and not plagued by the problems that windows is burdened with, or at least those problems that end-user clueless types consider to be important. Creating end-user apps for the platform where our end-users are is the very best way I can think of to gain insight into what they consider to be important. By ignoring windows as a platform for open-source development we're only helping Microsoft keep the barrier to use of Open-Source products artificially high.
Lee
Muslim community leaders warn of backlash from tomorrow morning's terrorist attack.
For christ's sake, slashdot, GNOME has had a new logo forever. Can you please update it?
One of the best features for Excel is UDF (user defnied functions), which is basically a Visual Basic Function which Excel can use.
I've never used this Linux version, but if it's got a programming interface (which is realtime debuggable) then migrating the code should be straightforard.
Ideally, VBA could be handled natively
If they change the format to be incompatible, that will probably won't buy them much time when OSS is incompatible, but then they're stuck with fully supporting the old format anyway. They can try to force upgrades to use the new format, but it will take many years to get rid of the files in the old format. In that time it just serves as a source of bloat and bugs in MS's software. That gives more reason to switch to an open format which is more consistant.
I can dream, can't I?
boldly going forward, 'cause we can't find reverse
I call that a new rendering of the old logo.
The whole open source movment (GNU etc ..) have to ask themself , what is the biggest gain for us , emulate Microsoft or trying to inovate. Emulating Microsoft formats is just futile, they just have to change file formats , APIs or standards to break years of efforts. I urge every one to try to think out of the box , you dont have to be MS compatible if you have an offer thats really good. If you strive to be compatible you will end up with a second rate , somewhat compatible product that's best can be described as "it kind of read/write MS files" ....
See here
Yes, let's stucture our entire education system around computing so people will use Excel properly!
Many (most?) people don't want to know more than how to browse and 'do email' so trying to force-feed it to those people when they are kids isn't going to help. Besides, it keeps 'us' in a job.
/.: why the hell am I here?
Uh... one of the few reasons why I use Access is because of the pretty reports it generates. How do you replicate this with Perl?
"Would it kill you to put down the toilet seat?" -- Maya Angelou
I'm curious to see how Gnumeric performs in terms of its implementation of certain complex functions, such as Solver. Anyone know where we can dig this information up?
I know alot of business that are not going to be ransomed to Redmond anymore, they will either just keep using their old copy of Office, or switch to Star Office or retrain their staff to use simple linux desktops. Gates is about to get what he deserves a consumer revolt from his best customers the little guy. Thats why the adverts for server 2003 stress how cheap it is, ya right! Until after you actually buy it. It will be the same thing with the site license rental scheme coming for Office 2003 get them on it then stick it to them when they try to put it on another desktop. Total bullshit, and a complete ripoff.
The fact that you have to use MS office to communicate, and send financial data and reports to other businesses should be grounds for a monopoly break up of the Microsoft cartel. The fact that it has not happened simply means that they have far too much pull with the US Republican government, and need to be chastised by their customers in the same way their monopoly took out IBM, Digital Equipment, Corel, and many others. Next on the list is Adobe, that is the heart of the .NET strategy.
How many people are out of work because of them? MS is getting ready to outsource coders also, just look at the big picture MS has caused depression, cutbacks and layoffs everywhere except Redmond, and Washington State. Not to worry Gates MS and Redmond either way you are next. Because of the Gates inspired .NET stupidity.
OH THE SHAME I fell off the wagon and use sigs again!
Once you define yourself as a competitor, then you can start adding the cool stuff that differentiates your product.
MS knows this as well. Excel just didn't materialize from thin air. Spreadsheets started with Visicalc on the Apple ][. It was a truly innovative program that, to the people who understood it, justified the purchase of the machine. In much the same way that the graphics capabilities justified the purchase of a Macintosh, even if it had barely enough memory. The one truly imaginative thing MS has ever done is was combine the spreadsheet concept with the Macintosh concept. The original Excel was a truly beautiful and a deserving successor to Visicalc. But Excel was only a successor, not an original. And since them MS has lost the beauty in a bunch of extraneous crap.
I cannot say the same thing about word, as MacWrite was a superior product for many years.
"She's a scientist and a lesbian. She's not going to let it slide." Orphan Black
I can't seem to find the Win32 binary of this program to try it out, (I have WinME). Where can I download it? Does it need IE installed to run? I'd really like to try Gnumeric out.
I just tried out a copy of Office 2003.
At work I use Office 2000, and use Excel a lot, etc.
I was previously using XP (2002) at home, and I noticed that there wasn't anything added on to Excel, or really to anything, just made it more "prettier".
The same is true with 2003, save Outlook which has been revamped.
It seems as MS is insisiting on keeping the same things. I know there are things here ad there that are updated, but nothing that would make you want to upgrade over 2000, and that's pretty sad.
GeekWares - Buy and Download Today!
=ROUND(SUM(A1:A10))
Do you really care if it's identical or not? The graphs are already good. If you need really fancy graphs, just use the free DX package, which does 3D, topographical and all that. The rest of Gnumeric works just the same as M$ junk except it's stable. Quit raining on the parade, this is great news.
I used to be a big Excel user, and thought it was the only thing Microsoft ever made that was worth anything. The trade up to free software was easy to make. On one hand, I had viruses and sytems that needed rebuilding every few months. On the other I had an adequate spreadsheet, excellent graph programs, compilers to make my own statistical analysis programs, and none of the Microsoft headaches. Excell and it's prety graphs are not worth the hastle of dealing with Microsoft bullshit, and though I miss some things, I'm much happier here. You, of course, will have to make up your own mind.
Here is great praise to the Gnumeric team! Your sheet is great for what a sheet is good for, simple small data set calculations. Keep up the great work.
Friends don't help friends install M$ junk.
i remember reading something on the mailing list about it...
you chump, its a new design.
Its a question of resource allocation. For any time we spend polishing our xls import export we get access to the installed base of MS Office. Whereas time spent on OOo's formats yields a much smaller number. Given more resources we'd spend more time on it, but for now I threw together an importer in a couple of weekends, and have only received one bug report in 6 months. given the documentation for the OOo format, and more importantly the existance of readable code that impliments it, it would be a simple project for someone to improve our support.
Even if it does some things better than Openoffice, it doesn't have a fraction of the momentum, the public visibility. It's kind of like Linux vs. *BSD.
Wait a minute! I could swear that I recently read that BSD was dead! In fact it might have been right here on slashdot!
After all, when you learn Linux, you already have to learn Shell, Perl, PHP, etc. [non-MS stuff] just to keep UP and be useful why make every single app do it's own thing rather than build soms synergy between them. I don't think Linux needs a VB clone. MS has already killed it. RIP VB!
I downloaded and Installed it. I decided to stress test it with a large excel spreadsheet (Around 1500x10x3). It imported all the data ok, but it ruined the fonts! I had to change it back into courier when it had changed it to arial.
Its Interface is still a bit confusing and is filled with engrish such as "replace by" in the search and replace dialog, it should be "replace with". I think the "HIG Police" should come after Gnumeric.
The Insert Image dialog does NOT have a preview pane (which IS ESSENTIAL WHEN CHOOSING IMAGES, I don't want to accidently insert the goatse guy!).
They Messed up the zooming dialog, now its harder to use.
GNUmeric is so close to beating excel, but the little issues stop it from beating the "doe test".
Also, how about some integration of OSS programs! The GNUs, K's, and OO.orgs office programs should all be able to read and write each other's native formats when possible. The current amount of division is silly and childish. There have developed 3 seperate-but-equal Linux "paths". The GNU, KDE, & Alt [mozilla/oo.org/etc] There should be an attempt make to keep things talking between the paths. i.e. files created with gnumeric/abiword should be equaly readable with Kword/Kalc & OO.org.
We support wk1, and wk2. There are some specs available for newer versions but I haven't had time to do more work on it. This sort of development is very parallizable. The i/o plugins are completely modular, so if anyone is interested in a new or esoteric format its not at all difficult to get something in place.
If anyone is interested please contact me.
You left Lotus123 out of your "history".
And we can read their encrypted files too, _the horror_.
:-)
Thankfully there's not much to worry about here. A few years ago MS published a series of 'MS Excel developer's kit' books. There wasn't much in the way of useful material so for actaully developing extensions to MS Excel available. So, as far as I can tell, Microsoft tried to pad the book with what it figured was some marginally interesting filler, a full set of docs for the xls file format
When I bought the book I was quite irrate at the lack of useful content. It warms my heart all these years later to actually be able to put it to good use.
PS
They also included the full content of the book in various MSDN discs, and on their web site for several years. Then mysteriously pulled it a few years back.
I find Excel's graphing system to be incredibly awkward and inconvenient compated to just about any graphing application, and the graphs tend to be ugly. But I find Excel's Solver incredibly useful and flexible, and have been able to use it to do complex fits of equations to multiple data sets with some parameters common and some separate.
Actually that capability was one of the first things that went into Gnumeric. Its an area we trounce MS Excel on several levels. - Adding a plugin to gnumeric is trivial. Indeed the vast majority of the functions are in plugins that are demand loaded. - plugin functions can be written in C (like an XLL but with a much cleaner interface), python, perl, or guile Docs for python here
How many of those deals have you done? Excel is not used for the actual calculations, some nasty propriatory thing made by a crazy former LSU person is used to do that. The silly $60,000 program's results are looked at by Excel, but I suspect anything could do that. The actual calculations made are some rather funny money, and only Federal regulations could lead to such a Byzantine calculation. I have been aproached to replicate the thing and I just might.
You would be nuts to try to match those calculations in a general spreadsheet. You are correct about precision in integration. While the numerical methods are well known, a general program would be geared to Engineering accuracy in the results, within 5% because that's what you can measure. When greater accuracy is needed, it's time to turn to a special program or write your own. In either case you will have to prove your specific set up against test cases and known results. Bond calculations and other accounting programs that have to be accurate to the penny are good examples of where you need a special proven program. Other exapmples are nuclear criticality calculations, space craft thermo-dynamic work and radiotherapy dose estimations. A spreadsheed in cases like that is good only for independent verification of a proven system. The system is known good, but it's nice to verify each and every calculation with something completely unrelated and much simpler just to make sure everything is working.
Ah! a little web search shows what a clever troll you are! This letter from way back in 0.7 days promisses arbitrary prcision will come when "Gnumeric is caught up with Excel in other ways". Well crovira, you want it, get cracking and merge BC into Gnumeric so that I can have arbitrary prcision by point and click. -crack- feel my whip and get coding bitch! I'll just keep using the older tried and true method of special programs for special applications.
Oh wait, this fsf thing from year 2000 says use guile. Damn it, I'm going to have to read the article. Nope not there. The world of free software may have given you the bizare and strange thing you want, but I can't find it. Start writing and let us know when you are done =;)
Friends don't help friends install M$ junk.
Our stated goal in the Gnumeric project is to produce the best spreadsheet available. After some consideration we decided that Gnumeric seems likely to produce that sooner than OpenCalc. OO is an important project, but as a spreadsheet user I have little interest in an office suite. There are quite a few users out there that seem to have similar views.
This is not a winner take all situation. OOo is the right solution for some users. However, Gnumeric is better is several areas already and with some work, we'll move past Excel in more places too.
If you want an Office Suite, by all means use OO. If you want the best possible spreadsheet I'm guessing that people will end up using Gnumeric.
That's really nice and greatly exceeds my expectations for a spreadsheet. I would not, for instance, expect a graphical representation for something that derives the mean and standard deviation of thousands of data points. Back when I was doing data analysis, I simply wrote a program from scratch to manipulate, or simplify by averaging many points to one, my data like that and dumped the results into a sheet for inspection and simple graphing. Then again, perhaps my expectations have been set by crappy closed source software.
A brush with a trolly post got me thinkng about my expectations for a sheet. Why not have more fuctionality in Gnumeric than Excel? There are all sorts of cool things that have been done in free software that can be brought into gnumeric. I can imagine arbitrary precision, arveraging "wizards" and modules that manipulate data files before displaying the results. The possibilities are exciting, and I'm almost ashamed to have been happy with good ASCII parsing, multiple formats, excellent autofill and all the other things 1.0 does. The more I think of it, the more I want to contribute, but my situation is not stable right now.
Thanks again for the great work.
Friends don't help friends install M$ junk.
I want delaware!
Nice work demolishing that troll, keep on going! Yeah!
It is possible to emulate, so that the user of M$ junk is not lost in the interface or data export, yet innovate by providing more and better functions. The world of free software is vast and includes routines for all sorts of math works. There are free libraies for arbitrary precision and forrier analysis for example. Why not have a nice little face on them in a free spreadsheet? Microsoft's limitations are not a straight jacket for the free world.
Friends don't help friends install M$ junk.
Ugghhhh. Yuk! I want to puke. FOR REAL.
Please Mod ALL Responses by Ms. Goldberg UP!!
As she's the Developer POC of Gnumeric, it should be prudent for us to see her response to our questions.
She does appear to answer every legitimate comment/question we're posting; and I'm getting a bit tired of looking down the thread tree for her responses.
Thanks.
Microsoft will change the rules of mathematics.
-CausticPuppy "Of all the people I know, you're certainly one of them." -Somebody I don't know
Why would changing the file format be detrimental to their own goals? They want everyone to run out and buy the latest version of Office, right?
A colleague who used to work for Microsoft told me that the people working on the office software always extended the file formats each release to force upgrades. They would always provide options for exporting to older formats, so as not to be so blatant. But they make sure that it wouldn't work about 1% of the time. So the user with the older version would eventually get fed up with the occasional failure (and with always having to ask people to send in the right format), give up and upgrade.
I agree but I have a tendancy to rant on these particular topics, because someone has to bash the machine in detail! :-) ...- Happy trails.
OH THE SHAME I fell off the wagon and use sigs again!
The ad sez that Gnumeric support _all_ Excel functions.
But where's that Pivot Table function that not only Excel had, but also Lotus Agenda ?
Muchas Gracias, Señor Edward Snowden !
Just wondering why Pivot Table isn't counted as one of the function of Excel ?
I mean, the ad sez that Gnumeric can now do _100%_ of Gnumeric functions, right ?
How about Pivot Table ?
Can Gnumeric do Pivot Table ?
If not, why not ?
I really hate to say this, but to me, 100% means 100%, and anything less than 100% shouldn't be called 100%.
Otherwise, it would be deemed as a practice of
deceptive advertising.
Muchas Gracias, Señor Edward Snowden !
You don't need Excel 2003 to know that Gnumeric doesn't have all the Excel functions.
For example, the Pivot Table function is _still_ missing.
Yes, the ad sez that Gnumeric has _all_ the Excel functions, but it is a misleading ad. If the Pivot Table isn't there, then Gnumerica can't have _all_ the Excel functions, period.
Muchas Gracias, Señor Edward Snowden !
Maybe MS should rewrite the GUI and WIN applications to run on the LINUX kernal.
-seems like the only good way MS might produce a good 64 bit windows.
Sorry buddy. You're really looking for Open Office, then, which can sort of understand it. As long as the charts and attached memos aren't too complex (or rely on some weird OLE control), it can usually import them.
Gnumeric is just for straight spreadsheet work (and it's damn good at it too).
THIS THING CAN TURN ON A DIME, MACROSSZERO STYLE ALSO FUCK BETA, ~NYORON
What does "-2^2" equal in Gnumeric? Excel claims it's 4 (as in (-2)^2) which is VERY wrong, as "-2^2" is actually "(-1) * 2^2".
We do not live in the 21st century. We live in the 20 second century.
since OpenOffice.org is moving to an xml document system, and there was once talk about having an open source document file system, so that different office-type apps can share them easily(like rich text docs, but richer ;) Couldn't OpenOffice and Gnumeric have a file that they both understand and kinda of lower the importance of haveing MS office vs OpenOffice vs Gnumeric?
Ofcourse OpenOffice has more publicity, but projects like OOo and gnumeric and abiword fighting over such a small userbase seems silly, when they should be working together to take over the MS Office userbase.
Floating point math can only be trusted when you take into account the effect of round-off errors on the significance of the result. So, yes, ((1+(1E-16))-1)*(1E+16) can be zero in Excel. (and probably Gnumeric too.)
If there are any real (non floating point roundoff/underflow related) errors in Excel, I'd really like to know about them.
Be wary of any facts that confirm your opinion.
"What's your hourly wage?"
Don't be a dolt...if you already have MS Office, there is no difference in price between this and MS Office.
If you don't have it and can't read a customer's spreadsheet for 1 day, you've paid for MS Office.
The fact that you're working at Burger King for $6.00 an hour doesn't mean that some of us aren't making 6 figures a year and can't afford to fuck around for a day for a political agenda.
Honestly, grow some common sense.
I think by "functions" they meant as in the programming sense of the word: the things you can type into an equation to do something funky. I don't use Excel, so my example isn't very inspiring, but think SUM(), AVG() and so on. Supporting all of these means that Gnumeric can load Excel spreadsheets without having equations it can't deal with.
For the majority of home users, a scripting language isn't needed. But for complex corporate sheets created by spreadsheet experts, a scripting language is necessary. If they have a large library of VB code in their spreadsheets, you'll never get the corporate people to switch, given how much time and testing they have devoted to developing these solutions.
I'd like to see a better scripting language than VB (python, perl, tcl or all three!) but there are market segments who will reject Gnumeric out of hand without painless import of VB macro code.
And thank you, Jody, for taking the time to answer so many questions on this thread. I find the best threads at Slashdot are those where the authors or others in the project take the time to address and answer the questions of Slashdotters. It really makes the thread worth reading, especially where you outline your design choices and plans for future implementation. Nice work all the way around.
Gnumeric could get me away from Excel right now except for VB macros. But with the increased support for scripting upcoming, I might switch anyway.
Your plans to release Gnumeric on Win32 should receive a high priority. Don't let Excel have the entire field by default. Gnumeric already can toast the silly Works spreadsheet and compares very well with Excel. Here is a prime opportunity to hit Microsoft where they live, right in the money tree and on their own platform. Currently, when friends/clients find that Works isn't enough for their needs, I have to tell them that Excel is their next choice. Gnumeric on Win32 can change that. I just don't find the OpenOffice spreadsheets very compelling. Gnumeric is just better.
A Win32 Gnumeric could be a nightmare for Microsoft and could further the growing dialog among sysadmins who advocate for Linux and OSS solutions in the corporate environment. You could knock a big chip out of the MS Office monolith with Win32 Gnumeric.
Give me that and I'll be happy.
News flash ... not even EXCEL can load all other EXCEL spreadsheets. The bulk of the spreadsheets out there are doing basic accounting and data sorting - I learned to use Lotus123 to do parts lists, notcash flow. Very few businesses I have been in have the customised VBA "workbooks" and office integration that Microsoft sells their MSOffice Developer version for - and at least one of those was rapidly changing over to web-based CGI or JAVA apps. I've used that software and it's no fun having VB fuctions that don't work like the examples in the book, or work differently when called from within a loop than they do inline. (I don't remember which one that was, but it was a PITA!)
Many (most?) people also don't want to know how to add, solve a quadratic equation, spell "ridiculous" and "then" properly, learn the history of their country and their political system, how they evolved, or to speak another language.
Does that mean the education system should not try to teach them these things?
Democracy is a horrific way to run an education system.
"I think we should learn how to tax dodge in school!"
"I want my sons to know how to get away with a bit of date rape!"
Computing is a pervasive aspect of modern life. To leave it to the few people who can be bothered to seek it out is purposefully lessening our ability to progress. The level of computing taught in schools is excremental, they are taught as if a computer is an appliance - not a general problem solver.
The point is not to get people to use spreadsheets better. Its to get them never to consider spreadsheets in the first place, as they are very rarely the best tool for anything even mildly complex.
"Gnumeric successfully copies Excel in at least one area"
You may be surprised to learn I'm not trolling. I'm trying to make a point. The way to get people to switch to open source software is not to announce you're now less feature lacking than a Windows product than before. "So, what else am I missing out on? Maybe you do get what you pay for..."
Well, I actually meant that more in a mind-trick kind of way - it's all about conception. But anyone using VBA in Excel is lost - and not only to other spreadsheet apps ;-)
Lars T.
To the guy who modded me down from perfect to terrible Karma - Apple haters still suck
I consider that a feature.
You can't judge a book by the way it wears its hair.
Why not write a PHP/Python/whatever module for MSOffice, and distribute it under BDS, then include the SAME API set in the various OSS products?
How many of you that use MSO would code your scripts in PHP, python or other OSS scripting language if it were avalible?
I"m confused as to how the only reply to a comment can be considered redundant. Any moderators care to respond?
"The meek shall inherit the earth, the rest of us shall go to the stars." Isaac Asimov
Compared to ms office and OOo, gnumeric offers very limited functionality to make printed output look nice.
E.g. I've just tested a single cell containing "dfsdfsdf", gave it a border and a background color (no drop shadow function available), and printed. Rather poorly, gnumeric didn't bother to render the right cell border.
This sort of thing may not be that important when calculating figures, but in the office world, crappy output is a show stopper.
gradually open source programs will get better at talking to each other but there is only so much time in the day and developers tend to place a higher priority on making their application better of itself rather than working on improved compatibility.
Help testing the SXC support in Gnumeric would be appreciated, Jody has only gotten one bug report so far.
.. the comment I referred to, earlier...,
but the example you showed me might well be "optimized" rather than calculated, eh?
-shrug- .. about a query that took over 20 minutes on Excel, but a minute or so on Linux: oughtn't they too be duplicated, even if one has to set an option for 'em ( like Stupidly-Bad Slowness Option, and Compromising Data-Integrity Option )?... /., where someone had a problem with Excel: they personally had switched to Linux/OSS, on their desktop, and they did a spreadsheet in their system, and their boss had a different result, so they went over every last thing, and discovered that the Excel version was rong...
I'd been thinking of defects, flaws, and other things that give Excel it's Look & Feel[tm], though, like the other comment ( wherever it is, I'm not going digging again )
hmmm..
I seem to remember another discussion, here on
I don't use Excel, so I'm stuck with the choice of gnumeric ( SuSE 8.2 pro installed the "stable" version, and it was obnoxious enough in interface & habits that I didn't like it -- now that I know-about the "development" version being good-enough for beta, by gnome's coding-standards, I'm downloading it to try compiling/running it ), KSpread, and OO.o's spreadsheet ( I don't much care for OO.o, as it's a resource-hog, and takes a very long time to start on my 5-year old machine, even with 320MB/K6-2 )... KSpread seems a tad unfinished, so I simply chose to work on stuff not needing spreadsheets for a few months...
Now I need one, and gnumeric 1.1.x awaits, so I'll dig into it ta see if I dig it, see..
I find the reports about bugs in Excel's accuracy to be probable, though, because MS cannot possibly code by test-first ( and end-up with the results they do ) XP-style, and almost all of the Excel users don't check the results with a different spreadsheet to see if the results are correct ( who backups their system? who checks their tools' integrity? who verifies anything ), and since it isn't OSS, bugs are more likely to remain long-term ( the stunning amount of serious, been-there-for-years vulnerabilities in windoze... )
Knowing gnome's coding-integrity standards ( not necessarily caring for the gtk interface, though: QT's much nicer to work-with ), knowing OSS's more effective in destroying & eradicating bugs, I'm not going to trust my future to MS-Excel, when gnumeric is an option. Period.
( thanks and generally-emanated hugs, for all gnumeric & gnome developers, eh? we gain vastly from this all, and appreciate it, though sometimes our griping-habit overrules wisdom & wellbeing )
Messages to/for me ( in me journal )
While MS Excel may have an extensive array of features it is somewhat lacking on the accuracy front. At least as far back as Sawitski (1994) various scientific analyses have been critising Excel using phases like "can be judged inadequate" and "it can be deduced that Excel uses an unstable algorithm". However as McCullough & Wilson (1999) note Microsoft has done little to address these concerns. The problems Sawitski found in Excel 4 were still present in Excel 97 and Excel 2000 for that matter. In fact critisism of the accuracy of Excel 2002 and XP in the scientific literature continues e.g. McCullough & Wilson (2002).
To quote the The Gartner Group, "Enterprises should advise their scientists and professional statisticians not to use Microsoft Excel for substantive statistical analysis". Of course if you do not need to do accurate statistical analysis then these problems will not effect you but given that Microsoft knows about and has largely ignored these problems and scientists are the people most likely to check that a given piece of software really does what if claims to do rather than using it blindly, it seems quite possible that similar problems exist in other parts of Excel but have yet to be exposed.
Rather than blindly copying Excel, the Gnumeric team might do better by trying to bring on board some of these scientists who have been testing and critising Excel in order to improve the accuracy of Gnumeric, so that not only does Gnumeric beat Excel on features but also, and far more importantly, on accuracy. See the following links for more info on the problems with Excel, 1, 2, 3, 4.
Again, I'd be happy if OSS woud at least work together to provide a "unified" choice to MS. I've been getting away from wanting OSS to have "just one Program", but it would be really nice if there were common file formats that were useful. [rtf & such don't count..they're 20 fricken years old! lets's get with the times] Sort of like MySql or PHP. They set such a standard on OSS that they were ported to windows...so you can learn OSS on windows, then switch!
On this we agree. A single unified front would be great. But, it will never happen. The strength of OSS is that every one does what they want and need it to do, not follow some decree or central set of ideas.
That said, I still think that we as a community should include php or other standard scripting language instead of arbitrary, app specific scripting languages.
And I think that if we could find a way to start allowing a few of these OSS scripting languages to find their way into MSO, it would make transitions easier when that time comes.
P.S. I like the idea of unified code. Access killer in the making?
Jody, thanks for the great piece of software! Gnumeric has done a lot for me, and I'm looking forward to seeing it become the gold standard of spreadsheets some day. Don't listen to the naysayers.
Secession is the right of all sentient beings.
And please stop with the Star Wars references and gay bear innuendo -- we get the point, okay?
When are you going to get in touch with the rest of us?
What the hell are you talking about?
Lars T.
To the guy who modded me down from perfect to terrible Karma - Apple haters still suck
Whoah! Don't you go waving your "angry wookie" at me!
Gnumeric seems to work ok on OSX. You need to install an X11 Server and Fink, and then, in a Terminal: sudo fink install gnumeric (and wait a long time) or sudo apt-get install gnumeric.
An Aqua or Carbon port would be cool though.
If more people used wxWindows, then (a) porting to Windows and Mac would be really easy, and (b) wxWindows would continue to improve as people contributed.
VOS/Interreality project: www.interreality.org