Gnumeric Turns 5

Jody Goldberg writes "Five years ago, Miguel committed the first code for Gnumeric to CVS. In a testament to the quality of the code several lines are still in use. Since that time the project has grown to more than 300,000 lines and now supports all 325 worksheet functions in MS Excel, plus almost 100 more. This seemed like a good time to thank all the people who have contributed to Gnumeric over the years. We're about to start the run up to the the next stable release which will be out in a few weeks and we look forward to continuing work with GNOME, and the community at large to produce the most powerful spreadsheet in the world."

Comments

Re:its time

[Anonymous Coward]

In case you didn't already understand that, this is just an imposter.

Gnumeric is great

[TheLastUser]

I use gnumeric all the time, I read MS xls files without any problems. Its also faster to start, and looks better, than OO (which I also like). Its my favorite of all of the Linux office apps.

Re:Gnumeric is great

[Anonymous Coward]

I absolutely agree. It is harder and harder to find a feature that I want to use that isn't there. The team seems to be focusing on the real beef of the application, making it really really good. Performance is a big priority too, try throwing a very very big spreadsheet at Gnumeric and you'll see how good it performs. And now that Graphs are (finally!!) getting there, there is very little reason to use another *nix spreadsheet.

Great job Gnumeric team.

congrats,
a happy gnumeric user

Re:Gnumeric is great

[BrokenHalo]

It is harder and harder to find a feature that I want to use that isn't there.

I can't remember who wrote the statistics add-ins for Excel (and I don't have a Windoze computer handy to find out) but that is one thing that would be very useful to me. Plus there is a range of plotting functions that are simply not there in Gnumeric, and I've been struggling along with Grace, which has a bit of a slow learning curve.. For all that, though, Gnumeric's a great product.

Re:Gnumeric is great

[Jody Goldberg]

Gnumeric has significantly better statistical routines than MS Excel. Some are home grown, others are from R.

Our charting utilites still have alot of growing to do. Hopefully with the new framework in place now we can start to accumulate features to target something like grace/xmgr

Several lines are still in use

[Anonymous Coward]

Comments with the authors name?

Re:Several lines are still in use

[ComaVN]

int main(int argc, char **argv) and ++i; come to mind too.

Re:Several lines are still in use

[Khazunga]

Argh, they said it was good code. If it were, it should read:

int main(int argc, char* argv[])

Re:Several lines are still in use

[Ben Hutchings]

Why do you think that is superior?

Re:Several lines are still in use

[Khazunga]

It's supposed to be a joke.

Anyway, I teach C programming to university level students. From my experience, the concept of an array of char pointers is easier to explain, than the concept of a pointer to a pointer being an array of strings. So, in a way, char* argv[] is better, although it's has the exact same compiled output.

Re:Several lines are still in use

[Ben Hutchings]

When declaring a variable, char *[] is not the same as char **. The odd meaning of char * argv[] in a function declaration is there for ancient compatibility and I suspect it will only serve to confuse the novice. main does not receive an array of strings, because arrays can't be passed by value in C.

Re:Several lines are still in use

[Khazunga]

When declaring a variable, char *[] is not the same as char **.

Given a.c:

void a(char* arg[]) {
}

and b.c:

void a(char** arg) {
}

and this set of commands:

gcc -c -o a.o a.c
gcc -c -o b.o b.c
strip a.o
strip b.o
diff a.o b.o
echo $?

We prove that char* arg[] is the same as char** arg, in the context of assembly code generation.

As for the correctness of using char** arg as opposed to char* arg[], the most I can say is that I never read of the usage of [] being deprecated in function call argument declarations. I'd thank if you if you'd give me some solid reference as to the deprecation. It was in wide use when I learned C, as can be easily observable in K&R.

Re:Several lines are still in use

[Ben Hutchings]

We prove that char* arg[] is the same as char** arg, in the context of assembly code generation.

Identical machine code proves nothing. You could change the int to long and still get the same working machine code on many platforms, but the program would be incorrect. However, I do understand that the two ways of declarating pointer parameters are equivalent. My point was that they are not equivalent when declaring any other variable. The meaning of [] in parameter declarations is an anomaly and potentially confusing.

As for the correctness of using char** arg as opposed to char* arg[], the most I can say is that I never read of the usage of [] being deprecated in function call argument declarations. I'd thank if you if you'd give me some solid reference as to the deprecation.

I never said it was deprecated, only that it's for ancient compatibility. Originally C did not have an array type, and arrays really were just appropriately initialised pointer variables. So it sort of made sense to use [] rather than * to indicate a pointer to array. Now that means something quite different in every case except parameter declarations.

Re:Several lines are still in use

[Khazunga]

I never said it was deprecated, only that it's for ancient compatibility. Originally C did not have an array type, and arrays really were just appropriately initialised pointer variables. So it sort of made sense to use [] rather than * to indicate a pointer to array. Now that means something quite different in every case except parameter declarations.

C goes back as far as the early 70s, but heavy C usage is a characteristic of mid to late eighties. By then C arrays surely were common. I should know, because it was by that time I learned C.

The common usage of * in array definitions is reminiscent of C-as-beautified-assembly. It assumes arrays are always allocated in contigous memory, and therefore pointer arithmetics are valid on array variables. If you think of it, this assumption is so prevalent it is now impossible to allocate C arrays any other way. The abstract concept of an array is in no way attached to contiguous memory allocation, and some other languages do in fact allocate resizable arrays in non-contiguous areas. This is why I prefere the [] form of defining array types. I'm digressing away from topic, though...

It seems to me you are trying to express that char* arg[] could be interpreted as the passing by reference of an array, since in parameters * means a pass by reference rather than by value. More, you seem to be saying that explicitely passing an array by reference is invalid, since arrays are always passed by reference. The interpretation would be correct, if a * in a parameter type only meant a pass by reference. However, it is a type modifier, creating a pointer off a base type. It marks a pass by reference as a side effect -- by definition, the pointer of a variable is a reference to the var.

As a type modifier, the usage of [] is as valid as that of *. They are both valid type modifiers, as of the ISO standard C.

Re:Several lines are still in use

[Ben Hutchings]

It assumes arrays are always allocated in contigous memory

That's practically the definition of arrays.

and therefore pointer arithmetics are valid on array variables

The whole purpose of pointer arithmetic is for working with pointers to array elements! The meaning of pointer arithmetic is undefined and generally unmeaningful except within arrays. As you know, an array can be implicitly converted to such a pointer and hence participate in pointer arithmetic; this implicit conversion is allowed for compatibility with the original implementation of arrays in C.

It seems to me you are trying to express that char* arg[] could be interpreted as the passing by reference of an array, since in parameters * means a pass by reference rather than by value. More, you seem to be saying that explicitely passing an array by reference is invalid, since arrays are always passed by reference.

C only supports call-by-value and pointers don't change that because they are created explicitly. So a parameter declaration that looks like an array declaration should declare an array parameter that's passed by value. Instead it declares a pointer parameter that's passed by value. If the function is called with an array then it's almost as if the array is passed by reference - which is inconsistent. Further, it isn't really call-by-reference - for example, sizeof(argv) returns sizeof(char**) not sizeof(char*)*argc.

Re:Several lines are still in use

[Khazunga]

That's practically the definition of arrays.

And here we hit the bullseye. You think this is the definition of arrays. I think it is not.

An array is a type construct which allows the programmer to deal with a set of equally typed variables. The set should be a mathematical set, in the sense that all the variables in the array should refer to the same knowledge domain. A programming language should provide an easy syntax for referring to singular members of the array by index -- be it a numerical index or a string index.

And the definition stops here. In fact, for 99% of the cases, the most efficient implementation is to allocate the array in contiguous memory. However, this does not cause this characteristic of most arrays to enter the definition of array. Except in C, again because C is a form of beautified assembly.

C only supports call-by-value and pointers don't change that because they are created explicitly.

No it doesn't. I defended this all along. Using char* arg[] I'm passing by value a reference to an array of unknown size. I think its valid under standard C, and you've failed to prove me wrong.

Re:Several lines are still in use

[Ben Hutchings]

The set should be a mathematical set, in the sense that all the variables in the array should refer to the same knowledge domain.

That's a very weird sense of "set". A mathematical set is an unordered collection with no duplicates.

No it doesn't. I defended this all along. Using char* arg[] I'm passing by value a reference to an array of unknown size. I think its valid under standard C, and you've failed to prove me wrong.

It's valid, and I never said it wasn't, but it's inconsistent with the way every other type is passed. If you teach people that this is a way of passing arrays rather than pointers then you're misinforming them. If you're teaching C then please make sure your students know how it really works. If you're teaching programming in general then why don't you use a different language?

Re:Several lines are still in use

[Khazunga]

I said:

I'm passing by value a reference to an array of unknown size.

You said:

If you teach people that this is a way of passing arrays rather than pointers then you're misinforming them.

If we're to reach anywhere in the discussion, please take the time to read my posts. You are taking the time to pinpoint small flaws in definitions and skipping the top level question:

• A program receives an a array of strings as a parameter. How best declare the parameter in C?

To me the logic is pretty simple. If the program received an array of ints, the syntax would be:

void a(int arg[])

a string in C is commonly represented as char*, so an array of strings is passed as:

void a(char* arg[])

There's no reason to treat strings differently. If the usage isn't deprecated, and if it produces the exact same object code on every platform (contrary to what you hinted previously), I'll stick to the most expressive syntax, thank you. One must have his mind twisted by C's preconceptions to automatically assume that char** arg is an array of strings.

I realize that something like int arg[] could mean that the whole array is passed by value, when it is passed by reference. However, jumping from that to using int* arg to represent an array is going from the pan to the fire. int* arg could as well mean one integer passed by reference. It's not expressive, and I recommend against its use.

As for the set definition, you realize that by scooping up the formal study book definition, you just shifted the domain relationship present in set elements down to the definition of collection. Anyway, the formal definition is not really important. I'm sure we agree on the purpose and correct usage of arrays. We just don't agree on the C syntax to refer to them.

Most annoying 'feature' of MS Excel

[civilengineer]

I use MS Excel almost everyday for data analysis, and the most annoying part is that number of records cannot exceed 65536 in Excel. Anyting larger than that, we need to get the data into Access and work in it, and that's not very fast and easy. What's the limit in Gnumeric?

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

Gnumeric is an Excel clone, therefore it has the same limit.

Re:Most annoying 'feature' of MS Excel

[OptimoosePrime]

now supports all 325 worksheet functions in MS Excel, plus almost 100 more

not exactly a clone

Re:Most annoying 'feature' of MS Excel

[Cid Highwind]

The same. (verified with gnumeric 1.1.19)
I disagree that it is an Excel clone though.

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

Unfortunately, I've found that it cacks after about a few hundred lines of formatted text.

I'm hoping that the new version is a little more robust for larger spreadsheets. Excel is still one of the applications which is tying me firmly to Microsoft and I'd love to go 100% Gnome.

Re:Most annoying 'feature' of MS Excel

[Jody Goldberg]

The default size is the same as MS Excel (256x64k). That helps ensure that all the funky xls files out there that depend on those limits work out of the box. However, those values are simple #defines. All coordinates are 32 bits internally. A quick edit and a recompile will change the bounds.

Re:Most annoying 'feature' of MS Excel

[C. E. Sum]

see the list archives.

The short answer is yes, but it's an issue that is being looked at.

Re:Most annoying 'feature' of MS Excel

[usotsuki]

I'll use gnumeric when it runs on XFree/Cygwin :)

BTW, I'd kill for a *good* macro system. I used to do Mad Libs in Lotus 1-2-3 2.01 macros on a 256K MDPA XT. :)

-uso.

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

That is the lamest excuse I have ever heard. What am I gonna do, recompile gnumeric for everyone in the office? Isn't the point of Linux that it's supposed to be BETTER than windows?

Re:Most annoying 'feature' of MS Excel

[schwap]

The default size is the same as MS Excel (256x64k). That helps ensure that all the funky xls files out there that depend on those limits work out of the box. However, those values are simple #defines. All coordinates are 32 bits internally. A quick edit and a recompile will change the bounds.

Why is this not a run-time feature?

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

Because of geniuses that rely on const char[] arrays.

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

I haven't checked the source myself... but surely they don't allocate a 256x64k block on the stack?!

Re:Most annoying 'feature' of MS Excel

[Jody Goldberg]

short answer :
I'm lazy

long answer :
Ya Ya I know. We'll get to it one day. Its not a terribly interesting problem right now.

Re:Most annoying 'feature' of MS Excel

[motox]

But unlike excel you got the source so its probably just a #define somewhere :)

Re:Most annoying 'feature' of MS Excel

[Rysc]

Gnumeric is clearly better than Excel. With Excel, there's a limit and you're fucked. With Gnumeric, there's a limit but if you really want to you can change it. That's a LOT better than Excel. THAT is the whole point of open source.

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

Cool. I will just get the secretaries to rebuild their own versions of Gnumeric from source. When they complain about how they don't understand I'll just tell them it's a LOT better than Excel.

Re:Most annoying 'feature' of MS Excel

[Anonymous Coward]

I don't think you get it. You can change the bound on Gnumeric but you can't on Excel. If you use Gnumeric out of the box it will have the same limits as Excel so what are you complaining about? If you really wanted to have a larger limit for the "secretaries" you could just make a package and use that on all the machines. Of course if all they have is an MCSE/you as a tech guy then I guess they're screwed. Don't blame Gnumeric because an idiot "techie" can't learn how to compile something.

OpenOffice

[k-hell]

How does Spreadsheet in OpenOffice perform against Gnumeric in terms of functions, compatibility with other spreadsheet programs ect?

Re:OpenOffice

[Maimun]

What about KDE's spreadsheet (Kspread, IIRC) and gnumeric? Last time I checked KDE's one, I found it inferior to gnumeric. But that was loong ago.

Re:OpenOffice

[bsharitt]

Both are quite capable as regular spreadsheets, but I think Gnumeric is still ahead as far as being the most feature complete.

Re:OpenOffice

[Anonymous Coward]

Ask Miguel. He created Gnumeric. He also stated emphatically that Open Office was better and that we should be concentrating on it, not Gnumeric.

I personally do like Gnumeric, but I also tend to agree with him when he said Open Office was "the most important open source project".

One thing we all know by now is without a great office suite the linux desktop isn't going anywhere. That and of course a calendar that can be shared to go along with OO.

Of course actually having desktop support and producst from companies like Adobe and Intuit would help, but let's start with the basics first. Like I said lack of a true Office replacement with easy-to-use Database and calendaring is what's keeping linux on the sidelines for all but a few companuies currently.

Re:OpenOffice

[axxackall]

I found OOo more compatible with recent MS Excel formats. But Gnumeric starts and runs way faster. So, when I don't care about excel files I always run Gnumeric. Functions I need are basically the same.

Re:OpenOffice

[Jody Goldberg]

If you've got any workbooks that we don't load properly please send a bug report. There is a confidentiallity protocol available if the data is sensitive.

Re:OpenOffice

[axxackall]

OOo Calc can save in Excel 97/2000/XP, which oftnen (but not always) cause less problems (than saving in Excel 95 format) for MS Excel 2000 to load it. Gnumeric can save only in MS Excel 95. I don't think it's a bug. It's rather a lack of a feature, specifically an absence of support of Excel/97/2000/XP format in Gnumeric.

I don't know what's the difference b/w Excel/95 and Excel/97/2000/XP formats, but if I can I would like to help the project from the QA side. Is a support for the Excel/97/2000/XP format in the development tree of Gnumeric?

Re:OpenOffice

[Jody Goldberg]

Yes, 1.1.x supports exporting to the newer 97/2k/XP formats.

Drop by the mailing list or #gnumeric on irc.gnome.org to meet the gang.

Re:its time

[Anonymous Coward]

-1 Troll? Praise indeed!

Really?

[nepheles]

This project could do with some marketing. I genuinely had no idea that it was even comparable to Excel in terms of features, and I'm no Linux n00b. One of the problems with OS software in general, I guess. And what has to change.

Re:Really?

[Anonymous Coward]

Its allowed to have bugs though because its still only version 0.00024251515 RC4 branch 6. just wait till RC5 in 2 years time and all will be well.

Re:Really?

[swv3752]

I know your trolling but to answer other people's question regarding Gnumeric, it works great. A solid project and a hidden gem of OSS. I hadn't realized that they had gotten around to supporting everything in excel. But the last couple of functions were so esoteric, hardly anyone used them anyways.

Re:Really?

[AvantLegion]

This project could do with some marketing. I genuinely had no idea that it was even comparable to Excel in terms of features...

Agreed.

My mother is an accountant and uses Excel for everything. She's generally not totally opposed to using Linux, except she does not enjoy OpenOffice's spreadsheet.

I've never heard of Gnumeric until now. So I'm currently emerging it on my Gentoo box, and if it looks good, it goes on my parents' computer's Red Hat install before the box gets sent back their way.

Re:Really?

[Khazunga]

--
You know Sergio?

Yeah!

Re:Really?

[Anonymous Coward]

Like maybe an article on Slashdot?!

Re:Really?

[follower-fillet]

> You know Sergio?
From Rio?

Interesting Software

[noldrin]

One of the things that have attracted me to Linux and GNU software in general have been interesting software such as Gnumeric. I hope the new office packages such as Open Office and K Office don't push out this type of software.

If Linux and GNU are going to get big, they have to innovate and write better software, not just emulate what the big guys are doing.

I want an office where I can use whatever software I want for each function, not what others decide to be in a suite.

Re:Interesting Software

[dekashizl]

I want an office where I can use whatever software I want for each function, not what others decide to be in a suite.

I do too, but often I can't afford to spend precious hours compiling and configuring just to get components to work together. Sometimes I'll take a hit in functionality just to be able to have things work out of the box. This is the appeal of "office suites".

Re:Interesting Software

[finkployd]

If Linux and GNU are going to get big, they have to innovate and write better software, not just emulate what the big guys are doing.

Interesting. So if Linux and GNU stuff perfectly emulated existing commercial (and expensive) software, which do you think would be big?

In many ways open source DOES innovate. Microsoft recently "discovered" Kerberos. IIS has been playing catchup with Apache forever (except in terms of conf wizards). Apple "discovered" BSD (well actually NEXT did, same diff). IE supposedly might finally have pop up ad blocking that the Mozilla has had for a loooong time.

I get your point, but I think you would agree that compatibility with existing popular applications is more important than new innovations. Once Gnumeric and Excel of equal feature-wise, then the race to innovate can begin.

Finkployd

Re:Interesting Software

[noldrin]

Both are important, that's why I like seeing the different software available for Linux. I just hope that Linux software makers try to work towards their software working together as much as possible. I'd hate for things to end up with Open Office being the one true office suite for Linux as MS Office ended up for Windows.

I feel that Gunumeric has innovated, and that's why I'm happy to see it getting press.

Re:Interesting Software

[Anonymous Coward]

Please don't assume that Miguel's inflated ego and cute little money-making scheme, aka GNOME w/ Ximian, has anything whatsoever to do with Linux or GNU. RMS might have adopted GNOME as the "official" GNU desktop, but that is a far cry from having anything to do with what GNU stands for. Infact, I'm not so sure RMS cares what direction the _software_ side takes--just as long as it's free software. As for Linux, well... it's just a kernel. And it's already big. Just today I read that Hitachi, Sony, Phillips and a few others are getting together and adopting Linux.

I want an office where I can use whatever software I want for each function, not what others decide to be in a suite.

Funny you mention OpenOffice and KOffice as pushing out Gnumeric. It was a mere 4-5 years ago that GNOME camp was going apeshit because of rumors that KDE crowd was going to incorporate GIMP into their desktop. My how times change and political spin works..

GNU/Linux/etc. are different things to each person that uses it/them. Never assume that we all want Linux to be a big desktop system. That is not my personal goal when I sit down to write software which I just happen to release under the GPL.

Re:Interesting Software

[Anonymous Coward]

Opera had pop-up blocking a longer time than a looooong time.

Re:Interesting Software

[alienw]

It also had really shitty HTML and CSS support for a looong time. It's still not great. At least Mozilla got rendering -- the most important part of a browser -- nailed down from the start. Opera is still playing catch-up there.

Re:Interesting Software

[WillAdams]

But by choosing Gnumeric you're letting other people choose for you 'cause Gnumeric is ``merely'' a clone of Excel done 'cause it's the most popular spreadsheet.

I'd give my interest in hell for more people in open source development to be more concerned w/ UI and to attempt to create better more interesting and innovative software---at the least, rather than clone the most popular thing (Excel / Windows), try the best (Improv or Quantrix / NeXTstep---for those who haven't checked in, www.gnustep.org has made some big advances of late ;)

William

Re:Interesting Software

[noldrin]

I'd don't recall ever saying what RMS or anyone else has for a goal for GNU or Linux nor do I recall saying that Open Office or KOffice has yet pushed out Gnumeric. I just said that GNU and Linux developers would do good to innovate rather then job copy as I believe brings more people in and helps create a buzz about a systems.

Re:Interesting Software

[Anonymous Coward]

Which is the problem. You assume too damn much. Who do you think is mostly in charge of GNU? RMS, of course! You assume that "GNU and Linux developers would do good to innovate" and I must also assume you mean exactly RMS and Linus. Other than that, no one gives a damn what _you_ want. I might write programs which fall under the GPL, but I'm not a "GNU" developer nor a "Linux" developer. I'm not in some stupid campaign to kill off Microsoft, either. Don't assume.

Re:Interesting Software

[salesgeek]

if Linux and GNU stuff perfectly emulated existing commercial (and expensive) software, which do you think would be big?
The closed source stuff. User base is entrenched and would see no reason to retrain. Many companies (to the chagrin of MS) still use Office 97 for this reason. If you want to displace a segment in the software market you've go to have a compelling reason to change:

* Highly desired enhancements of existing features
* New functionality - tools that give people new ways to use your package. Pivot tables, for example, allowed many business users to scrap crystal reports.
* Easier data interchange. Interestingly enough, MS Excel will talk to just about anything.

Remember, software's real value isn't measured in cool. It's a matter of if your package will let people get more done in less time with less work.

Re:Interesting Software

[finkployd]

Umm, you are not listening.

if Linux and GNU stuff perfectly emulated existing commercial (and expensive) software, which do you think would be big?

It would be fiscally irresponsible for ANY company to choose commercial software over open source (free) software IF it emulated the commercial stuff perfectly. The problem of course is that right now, it does not. There are still headaches using OpenOffice in an MS Office world.

What will make people think about switching first is the money they will save. MS Office ain't cheap, neither is Windows. I know of a few Universities planning a mass exodus from Windows servers because of the CALS license issue. The writing is on the wall and we all know that MS will only ever charge more and more for everything they possibly can. Perhaps Palladium is the early stages of a "pay per use" scheme, would you put it past them?

To switch (purely for financial reasons) the replacements have to support everything the existing stuff does. Once this happens, THEN the arms race to see who can really innovate will begin. Right now OSS is still playing catch up.

Finkployd

Re:Interesting Software

[salesgeek]

Actually, I did listen. You missed the real point: why am I going to replace the item I bought and paid for last year with a freeware clone this year?

Most software users already have bought the level of functionality that is being duplicated right now. There's no reason to invest time/money/energy into the opensource replacement. Many companies don't even see a compelling reason to replace the 1997 version of excel they have with Office XP.

My point is that for any software product to take a leadership position in the market you have to do more than imitate. I like some of the direction with Gnumeric - the functions, potential for python/perl scripting, etc...

On the idea of having to catch up, I disagree - you can direct energy at innovation at any point.

Re:Interesting Software

[finkployd]

why am I going to replace the item I bought and paid for last year with a freeware clone this year?

You wouldn't. That is why even if open source software was ready to take over the world tonight, it would still take time.

Many companies don't even see a compelling reason to replace the 1997 version of excel they have with Office XP.

Well, when Microsoft stops providing bug-fixes they might. Or perhaps it will be when the next version of Windows no longer supports Excel 97. Or perhaps they need to deal with external people who send them spreadsheets in MS Office XP format. I have never worked at a place that didn't keep current with software, in fact it is irresponsible to not do so, especially with internet software and operating systems.

It is in Microsoft's best interest to make sure they keep selling you Office over and over again. The primary "weapon" for this is intentional incompatible file formats, but I'm sure Palladium will play a role in forced upgrades as well. That is another advantage with open source software, they keep making new versions, but you don't have to keep paying for them. Unless the project is abandoned but that happens just as often (if not more) in the commercial world.

On the idea of having to catch up, I disagree - you can direct energy at innovation at any point.

Very true, but I would guess matching features with popular commercial apps is higher on the priority list for the developers.

Finkployd

Re:Comparing linux software to windows

[Anonymous Coward]

Nice, you have the "I expect to be modded down" element in your troll. That makes some stupid moderators reluctant to mod you down.

What are some of the extra 100 functions?

[Spoing]

I'm curious what was added beyond what is offered by Excel. Any really interesting little tidbits?

Plotting

[starseeker]

guppi is OK, but I really wish they would port the guts of something like Grace over to Gnumeric. One of the things MS did correctly on Excel was decent graphing, and I would like to see Gnumeric use some really powerful plotting tools from the open source world (scigraphica might be easier to incorporate) and take that to the next level.

Oh - is there any way to keep the scroll bar from reflecting the fact that there are 65000 rows or whatever in a sheet? It really limits the use of the scroll bar.

Re:Plotting

[Dashmon]

Use gnuplot.

Re:Plotting

[Anonymous Coward]

They should incorporate GnuPlot into Gnumeric. Then it could truly contend with Excel. Every time I've ever used Excel, I've also used its plotting functions.

Re:Plotting

[samhalliday]

oh dear oh dear, someone ALWAYS says it.. but gnuplot is not, err, GNUplot. if you know what i mean. it was started way in the early eighties and the name GNU is meerly a coincidence. the license is "open source", but it is by no means a license which allows the use of the code elsewhere in the world.

also, gnuplot is VERY hard to get it to look good. its the best, but it is really the graphing equivalent of LaTeX. you will never get that quality form a WYSIWYG graph program. havign said that, i so really think grace is a beautiful 2D plotting program (2D has gotta be emphasised, gnuplot beta just kills everything in 3D).

i remember having a discussion on the gnumeic mailing list about this very point, and they said that guppi gives a mass of power to graphing, which they have never even tapped in to. beyond the basics.

thats besides all the technical difficulty of getting 2 COMPLETELY different applications to talk with each other...

Re:Plotting

[Mr Z]

I recently integrated a macro preprocessing engine written in C++ into an ancient, pre-ANSI-C assembler that I didn't write. :-)

At least language wise, those two languages are just close enought to be dangerous. The styles were completely different. But I was able to integrate them by keeping the modularity sane.

To integrate something like gnuplot into gnumeric, they'd have to work on keeping the interface small, well defined, but still large enough to support all the desired functionality. Not impossible, but not a task I personally would envy.

Re:Plotting

[Cipster]

Or use R . . I find it better than Gnuplot.

Re:Plotting

[hankaholic]

One of the things MS did correctly on Excel was decent graphing

I'm sorry, but I have to strongly disagree with that one. I last really used a spreadsheet for serious graphing purposes about 6 years ago, using Quattro Pro (which came with Corel PerfectOffice) on Windows 95.

It had more flexible (and better-looking) graphing functions that Excel still can't touch, and was a damned sight cheaper too (I paid $90 at the time).

I'd say that Excel graphs like Pico edits text, but that's unfair to Pico.

Re:Plotting

[sketerpot]

From the gnuplot FAQ:

I wanted to call it "llamaplot" and Colin wanted to call it "nplot." We agreed that "newplot" was acceptable but, we then discovered that there was an absolutely ghastly pascal program of that name that the Computer Science Dept. occasionally used. I decided that "gnuplot" would make a nice pun and after a fashion Colin agreed.

Re:Plotting

[samhalliday]

aye, thats the one. cheers sketerpot ;-)

Re:Comparing linux software to windows

[rzbx]

Why is it bad to compare OSS software with a proprietary counterpart? I think it gives those that don't know about the software a chance to see how they compare.
If someone for example uses, MS Excel and wants to switch to the OpenOffice equivalent or Gnumeric in this instance, then they could see before hand if it contains all of the features they use frequently. At the same time it could show them features they have always wanted but could not get with the proprietary software. We compare things all the time. Is it really so wrong to do it with software?

Re:Comparing linux software to windows

[practicalista]

Don?t see any reason why this should be modded down. Surely the more interesting question is how many functions/features does Gnumeric have that Excel doesn't! Time for Open Source to stop chasing the pack, then the battle is won.

Yup.

[i_need_no_nick]

I'm sure the good people developing this will be glad that we've thanked them by melting their servers...

All well and good, but...

[dimator]

Why in the hell does it have to be so fragmented?


# apt-get install gnumeric
Reading Package Lists... Done
Building Dependency Tree... Done
The following extra packages will be installed:
bonobo-activation gconf2 gnome-mime-data libbonobo-activation4 libbonobo2-0
libbonobo2-common libbonoboui2-0 libbonoboui2-common libcupsys2 libfam0c102
libgal2.0-3 libgcc1 libgconf2-4 libgcrypt1 libgda2-1 libgda2-common libgnome2-0
libgnome2-common libgnomeprint2.2-0 libgnomeprint2.2-data libgnomeprintui2.2-0
libgnomeprintui2.2-common libgnomeui-0 libgnomeui-common libgnomevfs2-0
libgnomevfs2-common libgnutls5 libgsf-1 libgsf-gnome-1 libidl0 liblinc1 liblzo1
libopencdk4 liborbit2 libpopt-dev libpopt0 libstdc++5 libxslt1 libxslt1-dev
The following NEW packages will be installed:
bonobo-activation gconf2 gnome-mime-data gnumeric libbonobo-activation4 libbonobo2-0
libbonobo2-common libbonoboui2-0 libbonoboui2-common libfam0c102 libgal2.0-3
libgconf2-4 libgda2-1 libgda2-common libgnome2-0 libgnome2-common libgnomeprint2.2-0
libgnomeprint2.2-data libgnomeprintui2.2-0 libgnomeprintui2.2-common libgnomeui-0
libgnomeui-common libgnomevfs2-0 libgnomevfs2-common libgsf-1 libgsf-gnome-1 libidl0
liblinc1 liblzo1 libopencdk4 liborbit2
The following packages will be upgraded
libcupsys2 libgcc1 libgcrypt1 libgnutls5 libpopt-dev libpopt0 libstdc++5 libxslt1
libxslt1-dev
9 packages upgraded, 31 newly installed, 0 to remove and 485 not upgraded.
Need to get 12.0MB of archives. After unpacking 41.1MB will be used.
Do you want to continue? [Y/n] n
Abort.

Re:All well and good, but...

[GigsVT]

You apparently didn't have Gnome already installed.

Re:All well and good, but...

[Anonymous Coward]

Whine whine. Why can't I run Excel on my computer? I have DOS installed. In order to use it, I will have to install Windows, which will require installing several thousand files and taking up hundreds of megs. Whine whine.

Re:All well and good, but...

[jasonbowen]

Well... do you program? They need those libraries to function, it's a much better thing than writing all the functionality from scratch. Reusable libraries make programming large applicaions, or even small ones, much easier. You don't want to reinvent the wheel everytime you write a program.

Re:All well and good, but...

[incom]

I don't think Gnumeric is to blame. Here is all that gentoo would need:

incom@incomnia incom $ emerge -p gnumeric

These are the packages that I would merge, in order:

Calculating dependencies ...done!
[ebuild N ] dev-util/gtk-doc-1.1
[ebuild N ] dev-libs/libole2-0.2.4-r1
[ebuild N ] gnome-base/gnome-print-0.37
[ebuild N ] gnome-extra/gal-0.24
[ebuild U ] dev-util/guile-1.6.4 [1.4.1]
[ebuild N ] gnome-base/bonobo-1.0.22
[ebuild N ] app-office/gnumeric-1.0.13-r1

Re:All well and good, but...

[Anonymous Coward]

Why in the hell does it have to be so fragmented?

Welcome to the wonderful world of Linux on the desktop. But if you think this is bad, try install GnuCash from scratch.

Re:All well and good, but...

[jasonbowen]

Install a Windows application and watch all the dll files and ocx files that get installed fly by in the progress window. If a programmer had to reimplement all the functionality they needed from scratch everytime they wrote an application, programming would take even longer than it already does.

Re:All well and good, but...

[jasonbowen]

No... *sigh*, Gnumeric requires the same libraries no matter what distribution it is installed on. You distribution already had some of the needed libraries installed.

Re:All well and good, but...

[Anonymous Coward]

Install a Windows application and watch all the dll files and ocx files that get installed fly by in the progress window.

Yep, yep. Start the install program, sit back, watch just watch the progress bar and bang, it's installed. Compared with Linux where each distro seems to have its own incompatable package manager. Or compile from scratch and manage the dependencies manually.

I used GnuCash as an example, as good as it is even the developers say "If you decide that you cannot live with the version that came with your distro, and you don't have apt-get or a similar tool, then be prepared for a rather long and involved install/upgrade proceedure. Note that even with apt-get, some packages may still need to be installed manually."

Seriously, installing any of the major Linux distros is as easy as installing windows. Easier in some cases. I think the next major hurdle for Linux is making installing apps as easy as using InstallShield (or equivalent) Windows. And no, "apt-get" isn't good enough.

Re:All well and good, but...

[Anonymous Coward]

You quote YOURSELF in your sig? Let me guess: you wear a lot of black.

Re:All well and good, but...

[Anonymous Coward]

Good in theory, not at all the case in practice. Typically GNOME uses GNOME/GIMP's libraries. What about the countless libraries already installed on the system? I'm willing to bet that there are no fewer than 5 linked list implementations scattered throughout those libraries.

Libraries are reusable, but the current trend is to rewrite and claim ownership of. If a library does not work as needed, does anyone ever patch the library or reuse its codebase? Not usually.

I'm guilty of this too. I used to use GTK+. I decided to not use GList, their linked list implementation, simply because it was much quicker writing my own which had defined ("documented" or "understood") semantics. Later GTK+ changed and my program was severely broken. Not using GList meant one less thing I had to check and possibly fix. Not that GList should have changed at all...

Re:All well and good, but...

[NotQuiteReal]

FYI - Real Microsoft office 2000 SR-1 Premium 150MB (word, excel, powerpoint and who knows whatelse...) (acording to add/remove programs.) I have ALL the other Offices (I have MSDN subscription, $800 on eBay, thank you very much).

I have no idea how big "just" Excel is. Everything is so intertwined now...

Re:All well and good, but...

[damiam]

What's wrong with using libraries? Would you rather it all came bundled in some huge monolithic executable that ate up tens of megs of extra memory every time you launched it? You have apt-get, after all, just press 'y' and it'll install it all for you. It's not like it harms you in any way to have gnumeric use GNOME libraries.

Re:All well and good, but...

[jasonbowen]

The original point was about a Linux intall taking so many packages and comparing it to a Windows install, that point was refuted. I've had no problems with up2date grabbing the necessary dependencies when I've wanted an app I didn't have. Your example really isn't apples to oranges though you have a point. The reality is that any Windows distributor is throwing in all the libaries with their install program whereas somebody like the creator of GnuCash is relying on the user to make sure that they can run the program.

Re:All well and good, but...

[Anonymous Coward]

Someone already said it but the difference is what libraries you already have installed not what distro you use. I am almost totaly kde and gnome independent so when I "emerge -p world" I get much the same as the Debian user. So do both of us a favor and stop making us Gentoo users look like idiots.

Re:All well and good, but...

[Xabraxas]

Yep, yep. Start the install program, sit back, watch just watch the progress bar and bang, it's installed. Compared with Linux where each distro seems to have its own incompatable package manager. Or compile from scratch and manage the dependencies manually.

I don't see what this has to do with the discussion. Any package manager should be able to install just as easily. Sure not all of them are compatible with each other but what has that got to do with anything? If you don't want choice then go back to windows. Linux is about choice, not about replacing windows. I do believe it will eventually replace windows but that was never the goal so don't try to make it the goal of Linux.

Re:All well and good, but...

[Anonymous Coward]

The original point was about a Linux intall taking so many packages and comparing it to a Windows install, that point was refuted.

I think I was looking a little deeper into the original post. I not only saw a long list of libraries that had to be installed or updated, I also noticed they're maintained by several different groups. I used GnuCash as an extreme example of this; they require libraries from at least half a dozen different development groups. I completely understand they did it to reduce their own development effort, but as an end user, I just pray I don't have to install or update GnuCash from scratch.

I've had no problems with up2date grabbing the necessary dependencies when I've wanted an app I didn't have.

You've had better luck than I've had. I've been using the much revered apt-get and half of what I try to install has unresolved dependencies. Compiling from scratch isn't much better because they assume you have the right versions of all the development files for all the different libraries they used. I'm just trying to run a stupid program, not make a career out of it. And if throwing in all the stupid, piddling libraries lets me run the program, I'm all for it.

The reality is that any Windows distributor is throwing in all the libaries with their install program whereas somebody like the creator of GnuCash is relying on the user to make sure that they can run the program.

That's a very programmer's attitude, it's very human, and it's very lazy. In the extreme it's a screw the user attitude, "if the user is not smart enough to install my program, then I don't want them to use it."

A couple years ago that attitude applied to Linux distros too. It was difficult to install Linux and it seemed most (not all, thankfully) Linux users wanted it that way, but the major distros have virtually solved that program. Any major Linux distro is as easy to install as Windows.

I see the same situation now with installing and maintaining Linux applications. RPM and apt-get just aren't good enough. I'm hoping over the next couple of years Linux apps become as easy to install as Windows apps. Actually, thinking about it, they should be as easy to install as Macintosh apps. Hell, if you're going to copy someone, copy the best. Right?

Re:All well and good, but...

[jasonbowen]

I agree that a lot of work needs to be done for Linux to even approach Windows or Mac OS X on the desktop. I agree with your point about installing Mac apps, Drop the icon and your done, can't be simpler than that. This all being as it is, I only believe this will happen when one dominant desktop prevails. For Linux to even hope to challenge it needs to provide everything Windows and Mac OS X provide from the development process to the end user experience. Standards drive the acceptance of everything, people are not going to deal with, "well for that app you need the kde libs and qt and for this other one you want you need gnome libs and gtk+, and for...". If this continues, Linux will remain right were it is now with regards to the desktop.

Re:Cool google trick

[Anonymous Coward]

Just in case you actually thought this was real, note that the result is not actually a google response, it's a fake webpage. As should be obvious to nearly everyone.

actually

[Anonymous Coward]

The majority of slashdot readers connect from windows machines.

Sorry to burst your little bubble.

Not a Clone

[Jody Goldberg]

Our goal is to produce a great spreadsheet.

Compatibility with existing products is required for people to be able to transition. We already have significantly better analytics than MS Excel. Over time we hope to become a superset of it in other areas too.

Re:Not a Clone

[Anonymous Coward]

NOT THE REAL JODY GOLDBERG!!!

Just read here: Gnucash 1.3.0 Beta Released

Re:Not a Clone

[Anonymous Coward]

Looks like the real Jody to me.

Re:Not a Clone

[thrillseeker]

What features would need to be added to gnumeric run a spreadsheet such as this configuration aid at IBM?

Re:Not a Clone

[wmspringer]

One thing I didn't see on the gnumeric page - what are the system requirements to run it?

Re:Not a Clone

[vrmlguy]

I can't help noticing that several functions are flagged as "Subset" support. That would seem to mean that Gnumeric is still not duplicating all Excel functionality. Unfortunately, there's no mention in the man pages for each such function exactly what is still missing.

In fact, there are still many man pages still missing. Several that I clicked on came up with a "Page not found" error.

Re:Not a Clone

[Jody Goldberg]

Nice test case. That think took way too long to load. I'll have a look at why.

1) It makes heavy use of VBA macros. Supporting those is still a research project.

2) The button object importer is missing a few attributes. So the buttons are improperly coloured and labeled in gnumeric.

3) It uses EMF images. We don't have a display engine for those under linux.

Re:Not a Clone

[Anonymous Coward]

need glasses sonny? i can get some fitted for you as soon as I finish with my other adjustment all over your pretty face

5 years old code? Huh

[johnkp]

Five years ago, Miguel committed the first code for Gnumeric to CVS. In a testament to the quality of the code several lines are still in use

It's nothing. As a testament to the quality of the Windows sourcecode they keep seleral lines of code back from the early eighties in active use.

Re:5 years old code? Huh

[kurosawdust]

DOS is a tad more than "several lines of code"...

Re:5 years old code? Huh

[Anonymous Coward]

Like the error messages?

Re:5 years old code? Huh

[Anonymous Coward]

Nope. The GPL disclaimer at the top of the file.

Re:5 years old code? Huh

[Anonymous Coward]

Yeah, that's not a sign of quality, it's a sign of laziness.

Re:5 years old code? Huh

[Anonymous Coward]

someone tell that to Microsoft.

ORBit Perl automation

[mikeophile]

It's not quite ready for prime-time yet, but this is getting closer to being able to code your macros in Perl.

EDIT

[Anonymous Coward]

About half true...

4 Its unstable
The kernel is stable, its just KDE and GNOME that crash a lot.

8 It doesn't run Windows programs
There is software available to do this.

9 You cannot buy a computer with Linux
They sell them at Wal-Mart.

14 Un-american
Explain?

17 Anyone and their 14 year old brother can add (buggy) code
All code has to be approved.

Re:EDIT

[Anonymous Coward]

YHBT. YHL. HAND.

Re:EDIT

[Anonymous Coward]

http://www.yhbt.org/

Re:EDIT

[Anonymous Coward]

14 Un-american
Explain?

It's... COMMUNIST SOFTWARE!!1
Let the witch hunt begin :)

Re:Comparing linux software to windows

[Anonymous Coward]

Everybody compares a piece of software to the market leader in its field. Office just happens to have the best spreadsheet going right. Anybody who looks at any piece of spreadsheet software is going to compare it to Office.

Office just happens to be a Windows piece of software (and MacOS too, I think).

So no, you don't deserve to be modded as a troll, but do deserve to be modded a "misguided".

Gnumeric is ok, but not THAT hot

[curtlewis]

I used Gnumeric at my last gig since I had a Linux box as my main system. I had to launch it from shell since you need an environment variable set in order to write Excel files. That was kind of annoying, plus I had to dig around on the net to find out what that variable was.

Gnumeric is just enough spreadsheet to get you by. It's pretty streamlined compared to Excel, which is the polar opposite, a bloated piece of corporateware.

And for the humorous part of this post, lines still in use:

#include stdio.h /* Gnumeric is licensed under the GPL */
i++
void main();

Re:Gnumeric is ok, but not THAT hot

[Anonymous Coward]

Question: why prototype main? Question: why prototype main to be unconforming to the standard?

Re:Gnumeric is ok, but not THAT hot

[Anonymous Coward]

you have no idea what you are talking about.

how about trying to use it today . very few new people will be using it a few years ago

Re:Gnumeric is ok, but not THAT hot

[samhalliday]

I had to launch it from shell since you need an environment variable set in order to write Excel files

.xsession my friend, this is where this kinda stuff is supposed to go...

Re:Gnumeric is ok, but not THAT hot

[jasonbowen]

You could just set that environment variable in your shell init script. Personally I've saved many sheets done in Gnumeric as Excel sheets and I never had to do this. What is the name of this environment variable?

Re:Gnumeric is ok, but not THAT hot

[Jody Goldberg]

1) In early 0.x versions we required you to set an env var before you could _overwrite_ a file in an exporter that was under development. Which seems pretty reasonable for a development release. That requirement was removed for all exporters before the stable series (1.0.x) was release a year and half ago.

2) The lines are
Sheet *
sheet_new (Workbook *wb, char *name)
{
Sheet *sheet = g_new (Sheet, 1);
do stuff
return sheet;
}

Ok they're not exactly high art, but Miguel started this project, and I believe in giving credit where its due.

Re:Gnumeric is ok, but not THAT hot

[Anonymous Coward]

or .bash_profile, or .profile

Re:Gnumeric is ok, but not THAT hot

[Emil Brink]

Since it's cool to pick nits in the Famous People's code, given an opportunity, shouldn't that name pointer be declared const? It sure looks like an input parameter in the original code linked to in the article blurb. Heh.

Re:Gnumeric is ok, but not THAT hot

[samhalliday]

no... bash is not your login shell when you login to X. you want your WM to see the PATH and other envars; but what you suggest only lets the bash shell see it. unless you make your .xsession a login shell of course, which involves some jiggery-pokery ;-)

Re:Gnumeric is ok, but not THAT hot

[Jody Goldberg]

Ok, so I cheated a bit :-)
The modern version is indeed const.

Re:Gnumeric is ok, but not THAT hot

[jcast]

unless you make your .xsession a login shell of course, which involves some jiggery-pokery ;-)

I don't think so---it should work just to make the #! line say #! /bin/bash -l.

Re:Gnumeric is ok, but not THAT hot

[samhalliday]

not on all systems, (i find at least on solaris and generic gnu/linux) that .xsession is only sourced by the system file, so it doesnt matter what the first line has. unless of course you add "#!/bin/sh -l" to the top of the system file (a hack, and its what i do on machines i can); but not everyone has root access, so on my user-only machines i "exec bash --login .another_file" in my .xsession which is the jiggery pokery i mean...

Re:Gnumeric is ok, but not THAT hot

[jcast]

Ah, right. I was of course remembering that I knew I was successful in add #! /bin/bash -l to my gnome session script on my desktop---but I have root, of course. I didn't think about the poor non-roots out there.

Btw., would it work to say `. ~/.bash_profile' or some such in your .xsession? I wonder...

Re:Gnumeric is ok, but not THAT hot

[samhalliday]

it should, but i've had mixed success. but come to think of it, i did "source ~/.profile", when like you say, i should be POSIX compliant and do a ". ~/.profile" because the SUN /bin/sh is a POSIX shell, not bash. hmm... sam goes off to remove the giggery pokery form his recursive setup...

Re:Gnumeric is ok, but not THAT hot

[jcast]

but come to think of it, i did "source ~/.profile", when like you say, i should be POSIX compliant and do a ". ~/.profile" because the SUN /bin/sh is a POSIX shell, not bash. hmm... sam goes off to remove the giggery pokery form his recursive setup...

Glad to be of service.

Re:Comparing linux software to windows

[Jellybob]

I'm impressed... the post is currently scored at +2: Troll

Re:Comparing linux software to windows

[usotsuki]

(Score:3, Troll)

Ye gads!

LOL, what's up with Slashcode?

-uso.
I think it should be "Insightful", myself.

what happened to GNU Oleo ?

[Anonymous Coward]

Richard prefers it to the high priced spread.

Re:what happened to GNU Oleo ?

[Anonymous Coward]

I use it. I'd like to see gnumeric with a white-on-blue ncurses text-console interface such as midnight commander uses. Heck, I'd PAY for that.

Re:what happened to GNU Oleo ?

[Jody Goldberg]

- We have a cheesy oleo importer

- I'd love to see someone do a text interface to gnumeric. We have a full model-view-controller split, so its feasible to do it. /me lays down the gauntlet. Is anyone up to the challenge ?

Re:what happened to GNU Oleo ?

[Anonymous Coward]

And we need textmode abiword, be great for using over telnet connections to remote places

Re:Comparing linux software to windows

[Xtifr]

For one thing, most people reading slashdot don't care, they already use linux.

No, most POSTERS on slashdot want people to THINK they run Linux (and a few even do). Most READERS of slashdot run Windows.

Everyone bitches when people compare the other way around

No, "everyone" doesn't, although (this being slashdot), someone bitches whenever ANYTHING is compared to ANYTHING else. But if you take the bitching on slashdot seriously, you've got deeper problems than just a need to feed your inner troll. :)

Anyway, I am a Linux user. In fact, until very recently, I was the maintainer of ORBit (the CORBA layer of Gnome) for the Debian project. And I wasn't aware that Gnumeric had gotten as far as it has, and I found the comparison interesting. (Even though I don't think I've used a spreadsheet since 1998.)

where are the binaries????

[Anonymous Coward]

I've searched for Win32 binaries everywhere. Anybody?

Re:where are the binaries????

[amd-core]

Use the source, Luke ;)

Re:-21, flamebaitse.cx

[MalleusEBHC]

7 You have to tipe commands

I can see how this would be a problem for you.

Re:Cool google trick

[Anonymous Coward]

So you didn't see that he wrote "Google" you idiot.

And still can't do much...

[Anonymous Coward]

And still doesn't hold a candle to Excel. Free software still can't do desktop apps (from scratch).

Wasn't Gnumeric supposed to be some kind of showcase of the types of things that could be done with the GNOME framework? Very little, it seems.

Re:And still can't do much...

[Anonymous Coward]

Can't hold a candle to Excel? I noticed that you did not provide so much as one example to prove your point. Obviously you can't since you've never used the software.

Unintelligent babble only demonstrates your complete lack of intellect. Supply an example or shut up.

Re:And still can't do much...

[Anonymous Coward]

Well, I am using gnumeric on a daily basis and
I find it more convenient and faster than Excel.

There is no things that I need and are in
Excel but not in gnumeric ...

But can it do this?

[Anonymous Coward]

Excel specifications and limits

Worksheet and workbook specifications

Feature Maximum limit
Open workbooks Limited by available memory and system resources
Worksheet size 65,536 rows by 256 columns
Column width 255 characters
Row height 409 points
Page breaks 1000 horizontal and vertical
Length of cell contents 32,767 characters. Only 1,024 display in a cell; all 32,767 display in the formula bar.
Sheets in a workbook Limited by available memory (default is 3 sheets)
Colors in a workbook 56
Cell styles in a workbook 4,000
Named views Limited by available memory
Custom number formats Limited by available memory
Names in a workbook Limited by available memory
Windows in a workbook Limited by system resources
Panes in a window 4
Linked sheets Limited by available memory
Scenarios Limited by available memory
Changing cells in a scenario 32
Adjustable cells in Solver 200
Custom functions Limited by available memory
Zoom range 10 percent to 400 percent
Reports Limited by available memory
Sort references 3 in a single sort; unlimited when using sequential sorts
Undo levels 16
Fields in a data form 32
Custom toolbars in a workbook Limited by available memory
Custom toolbar buttons Limited by available memory

Calculation specifications

Feature Maximum limit
Number precision 15 digits
Largest number allowed to be typed into a cell 9.99999999999999E307
Largest allowed positive number 1.79769313486231E308
Smallest allowed negative number -2.2250738585072E-308
Smallest allowed positive number 2.229E-308
Largest allowed negative number -1E-307
Length of formula contents 1,024 characters
Iterations 32,767
Worksheet arrays Limited by available memory
Selected ranges 2,048
Arguments in a function 30
Nested levels of functions 7
Number of available worksheet functions 329
Earliest date allowed for calculation January 1, 1900 (January 1, 1904, if 1904 date system is used)
Latest date allowed for calculation December 31, 9999
Largest amount of time that can be entered 9999:59:59

Re:But can it do this?

[Anonymous Coward]

Okay, here goes
Undo levels : 20!
Zoom range : 10-500%
Largest amount of time : ~580000:59:59 (before it overflows)
Earliest date : 01/01/1000
Largest date : 31/12/9999
Colours in a workbook : Limited by avalible memory.
Row height: limited by the limits of the font used.
Ability to get shitted on by tubgirl : yes!

So thats 8 things its better at, and it equalizes with most of the other stuff.

Re:But can it do this?

[Jody Goldberg]

undo levels are configurable.

You can set the number of levels, and a logical size limit. undoing filling an entire column is more expensive than entering a value.

Most everything else is 'limited by memory'

Re:TROLL STORE!

[Anonymous Coward]

I am not familiar when the lemonparty phenomenon. Please elaborate.

Re:Comparing linux software to windows

[Anonymous Coward]

That is because you are one of the targets of the troll's bait.

Missing Component?

[__past__]

Gnumeric is OK, but where is the VBA equivalent?

I'm serious. People in the Windows world use Excel not only to calculate stuff, but as some kind of application platform. Personally I think that's stupid in most cases, but not offering it is even worse.

Maybe I just couldn't find it anywhere, but: Is Gnumeric easily scriptable? It doesn't have to be Excel or VBA compatible (in fact, about every other language would be better, IMHO), it doesn't need an integrated IDE with debugger etc. like Excel has, but the only thing I could find so far is a "plugins" directory containing .so files - that can't be it. Is there something better, and if so, why the f**k isn't it documented prominently?

Re:Missing Component?

[Anonymous Coward]

Cobra::ORBit

Re:Missing Component?

[Eneff]

wait... you're telling me that you're supposed to do corba to talk to this thing?

the point of vba is that it's easy.

Re:Missing Component?

[BigBadBri]

but not offering it is even worse.

Why is it worse?

If you don't give the option to people to do something that (by your own assessment) is stupid, perhaps they'll be less lazy and do things properly.

A spreadsheet is a self-contained application - you can provide entry fields, and recalculate - that should be the limit of it.

I've seen several 'Excel applications', and they all suck.

Horses for courses, and screw the lazy developer.

Re:Missing Component?

[Anonymous Coward]

python

Re:Missing Component?

[weston]

A few years ago I wrote a fairly complex spreadsheet (a simulation of the Monty Hall problem) with some Excel Macro/VBA functionality. My memory on this is a bit fuzzy, so I could be wrong, but I believe I tested it under Star Office and possibly Gnumeric as well to see if it ran, and I seem to recall it did.

Interested parties could test:

http://weston.canncentral.org/misc/Monty/GameSho wF inal2.xls

Some background can be found here, if you're not familiar with the problem. The Workbook has one sheet for playing individual rounds of the game yourself, which you do by selecting an initial guess and then a second guess (one button for each option under each door). The next two sheets let the computer simulate playing 1, 10, or 100 rounds at a time, one uses the "switch" strategy and one uses the "stay" strategy. If it's confusing (probably is), email me. I'm curious.

Re:Missing Component?

[dustman]

wait... you're telling me you have to use OLE2/COM to talk to excel? the point of vba is that i's easy...

"Native" OS X port?

[squiggleslash]

Does anyone know if there's any work going on to port GTK/GNOME apps like Gnumeric to Mac OS X, in a standalone fashion? (ie so they don't need X11)

Right now there isn't a free office suite for OS X that doesn't run under X11 and hence look like donkey, though there's apparently an unreleased beta of a native KOffice that uses Trolltech's new Mac OS X native QT toolkit.

Re:"Native" OS X port?

[Arker]

Does anyone know if there's any work going on to port GTK/GNOME apps like Gnumeric to Mac OS X, in a standalone fashion? (ie so they don't need X11)

Right now there isn't a free office suite for OS X that doesn't run under X11 and hence look like donkey, though there's apparently an unreleased beta of a native KOffice that uses Trolltech's new Mac OS X native QT toolkit.

I disagree with your 'donkey' comment, completely. My mac is themed as close to Next as I can get it, and with apple X11, it's hardly donkey. Donkey is stock aqua imhop, although it's not quite the most donkey out there. ;) If X11 apps could figure out where to put their menus, that would be nice, but I'm not going to pull my hair out over it.

I would dearly like to figure out how to get this stuff working easily though - preferably through the package manager. I can figure out how to download the source and compile it, if I have enough time which unfortunately I don't just yet, but a nice clean package install would not only save me the trouble, it would make it easy to get my friends switched too.

Abiword had a package, not exactly the perfect format but beggars can't be choosers - only it never ran. I saw they had a new version listed, went to try it out, and the file isn't found. :( A quick search on gnumeric just found some notes on how to compile it with fink and somebody trying to sell out of date CDs. Ooh, openoffice finally has a mac download, I'll try that out now.

Re:"Native" OS X port?

[Jody Goldberg]

A truely native port is possible, would be alot of work. I'd rather see a native version of gtk+ for OSX. That would help alot more projects.

How far are the various ports from being usable ? I've heard that film-gimp has been working on gtk-1.2, but have not researched it. Are there any gtk2 efforts in the works ?

Commercial reenactment

[Anonymous Coward]

Q: HOW DO A NIGGA GET ON DUH INNANET???
A: LIKE TOTALLY LIKE, WHAOAH DUDE LIKE WOW OMG.
Ding ding DING!
Now with Intel Pentium 4 processors.

aslfj;aslj asdfdlkfjsl lksdjflskjdlf lkjslkfjs lowering the caps to noncaps ratio

Re:Commercial reenactment

[A_RADICAL_TROLL]

COME ON MOD PARENT UP!!!!!

we need to have the facts you goddamn liberals

FTFL

[xmda]

Well, if you had FTFL (Followed The Fscking Link) I guess you could have found out for yourself: http://www.gnome.org/projects/gnumeric/functions.s html... :)

Re:FTFL

[Spoing]

Well, if you had FTFL (Followed The Fscking Link) I guess you could have found out for yourself:

I did. There are 100~ differences. Want to give me the highlights?

Re:FTFL

[Jody Goldberg]

Correct versions of several functions that are broken in MS Excel. We need to support their incorrect values and the right answers so that imported workbooks stay the same.

More Statistics
More Random distributions
Lots of financial derivative pricers

Re:FTFL

[Rude Turnip]

Jody:

The derivatives pricing functions look *very* interesting to me...but when I click on a function's description link on the Gnumeric web site, I get a 404 error. They could save my company lots of money in licensing fees from an Excel-based option pricing model that we currently use.

Re:FTFL

[Anonymous Coward]

The links on this page seems to work better
http://www.gnome.org/projects/gnumeric/doc /functio n-reference.html

Re:TROLL STORE!

[Anonymous Coward]

The group of gay elderly men used to be featured on www.lemonparty.org.. its gone now

Limited by availible memory.

[Anonymous Coward]

That's deceptive, since when your run Excel, you don't have any availible memory left.

Desktop-specific afiliation

[thelandp]

Okay this may be a little off-topic, but Gnumeric is a perfect example of what this comment is about.
I'm concerned that so open source apps written these days have names that demonstrate their affiliation with a particular desktop. Having names that begin with "gn" of "K" is a kind of flag waving that shows which desktop application framework was used (gnome or KDE).
Ideally the user should be able to (and usually can) run apps using either framework on any desktop. But when the name has "gn" for example, are they saying "well yes you could probably run it in KDE but it's a gnome app so maybe you're better off running it in gnome..."
Why is their so much tribalism? I think it's an important step in the maturity of Linux or Open Source in general to get to a point where the particular implementation (gnome or KDE) of any given layer (the desktop) has NO impact on other layers (the application) and so the title of the app should not even need to provide any hint of affiliation with a particular brand of in another layer.

Happy Birthday Gnumeric, looks like a great program. But as a user I don't think I should need to know about it's internal implementation thanks.

Re:Desktop-specific afiliation

[janda]

It's my understanding (I'm sure lots of people will correct me if I'm wrong) that you can't necessarily just compile->run apps between desktops.

If I recall correctly, it has to do with the gnome desktop using c/GTK bindings, while the KDE desktop uses c++/QT bindings.

Re:Desktop-specific afiliation

[Anonymous Coward]

"But when the name has "gn" for example, are they saying "well yes you could probably run it in KDE but it's a gnome app so maybe you're better off running it in gnome..."
Why is their so much tribalism?"

They're trying to say it's part of the GNU project, nobhead. That's what the G means - GNU Numeric - it's got nothing to do with Gnome. Do you think gzip uses Gnome or something?

Re:Desktop-specific afiliation

[Anonymous Coward]

Possibly. But on my system (Debian unstable) I run QT & KDE apps under Gnome, and GTK & Gnome apps under KDE just fine. Of couse, they still have their native themes and file controls, but they do work.

Oh, drag and drop, and cut/copy and paste work just fine too.

Re:Desktop-specific afiliation

[__past__]

You can always run all Gnome apps on a KDE desktop and vice versa. You can also run them on twm, or use a plain XLib app with any desktop. That's great.

There are several desktop platforms, and there will always be, because they have different goals (KDE tries to make the desktop as slow and space-wasting as possible, while Gnome's goal is to remove as many useful configuration options as it can while avoiding cross-app integration ;-). Has never been different, will always be that way. Only that, thanks to X11, we have interoperability between them. The world is great if you are a Unix user.

Re:Desktop-specific afiliation

[thelandp]

My point is still valid. Any GUI app whose name starts with gn is probably using the gnome framework, and any GUI app starting with K is probably using the KDE framework. But that fact the we as users need to even think about this distinction means Linux is being split into two separate incompatible groups.

Re:Desktop-specific afiliation

[Eneff]

well, you do need gnome installed to use it... bonobo and all. it's not just gtk.

Re:Desktop-specific afiliation

[dozer]

It's true that Gnome apps use GTK and KDE apps use Qt. However, GTK and KDE interoperate extremely well thanks to the efforts of freedesktop.org.

It may surprise you to hear that you do not even need to run a Gnome or KDE to use their applications. I'm running a blessedly clean IceWM setup and I still get to use Evolution.

Re:Desktop-specific afiliation

[digitalhermit]

I dunno. Sounds interesting on the front of it, but I certainly don't agree. First, gnumeric does run perfectly well under my KDE desktop. Second, if the developers did it for the glory of Gnome, then so be it; they are free to call it whatever they want. Now I've purchased a few RedHat and Mandrake distributions but have never sent anything to the Gnome foundation itself (something I hope to remedy in a month or so). If they want to call their excellent program ElephantAss I'm certainly not going to argue. In any case, you are free to name your menu shortcuts whatever you want. E.g., RedHat doesn't call their email client "Evolution" but "Email". It still runs Evolution, but the user doesn't need to know that.

Re:Desktop-specific afiliation

[Rock Ridge]

In gnome they are paying tribute to the great white father "Gnu," when they name something gn-something. What do you suppose is going on with k-desktop, k-games, etc.?

Re:Desktop-specific afiliation

[phantomlord]

I personally stay away from the K* apps because I don't want the bloat^ of the QT/KDE libraries installed in addition to the GTK/GNOME libraries. I've been happily using linux for years without ever having to use a KDE app (and thus install the associated libraries) and barring something groundbreaking, I expect it will stay that way for a long time.

Thus the G*/K* naming convention is handy for me. I don't need to download a couple meg pile of source code only to find out that it's a QT/KDE app when I try to compile it

^note, I'm not referring to the size of the libraries but I simply refuse to install half of another DE I don't use simply for the sake of a couple apps

Re:Desktop-specific afiliation

[KeyserDK]

And how much of your harddisk would that actually use? It's not like harddisk space is the most expensive these days.

Re:Desktop-specific afiliation

[Anonymous Coward]

Actually, no. G in Gnumeric really _is_ for GNOME. gzip has nothing to do with GNOME. It only predates it and the G is an affiliation with GNU, rather than GNOME. As far as I know, Gnumeric has absolutely _no_ relation to the GNU project. It would only be tangently related from the GNOME project which is the "official" GNU desktop.

Re:Desktop-specific afiliation

[Arker]

It's my understanding (I'm sure lots of people will correct me if I'm wrong) that you can't necessarily just compile->run apps between desktops.

Consider yourself corrected. That's completely wrong.

As long as you have the required libraries installed you can run gnome or kde apps with or without the gnome or kde desktop crap. I've used both regularly under Windowmaker, IceWM, etc. without using their desktops at all. On debian or gentoo your installation command will automatically grab the required libraries when you install. You don't need, for instance, to use or even have installed the windowmanager, file manager, panels, etc. that make up the 'desktops' to use the programs, just the .so libraries.

There used to be a few issues getting them to cut and paste properly between them, but I understand that's been fixed.

Re:Desktop-specific afiliation

[BrokenHalo]

names that begin with "gn" of "K" is a kind of flag waving that shows which desktop application framework was used

It appears to be so with KDE, but not Gnome so much: e.g. Evolution, Sound-Juicer, File-roller, Drivel, Nautilus etc etc...

In any case, most distros tend to install most of the more popular desktops by default; and unless you really have a tiny HDD, there's really not much point in overriding that unless you have very strong feelings about one or another. For instance, although I'm not personally a big fan of KDE, I still very occasionally use Konqueror to check my html or even more rarely when a page won't load in Mozilla.

Re:Desktop-specific afiliation

[SagSaw]

Bullshit. Pure, simple, bullshit.

Yes, gnome and kde use different widget libraries. Applications designed for the gnome desktop rely (IIRC) on the gtk libraries and KDE apps rely on the qt libraries. However, the kde and gtk libraries co-exists, so as long as you have the underlying libraries installed, gnome apps should run under a kde desktop and kde apps should run under a gnome desktop. In fact, I use gnumeric as my spreadsheet program and evolution as my e-mail client despite the fact that I use KDE as my window manager.

Re:Desktop-specific afiliation

[don.g]

The $app =~ /$g/i => GNOME and $app =~ /$k/i => KDE thing is annoying, I agree. But it's quite likely that gnumeric is named that way due to GNU rather than GNOME (like GCC).

The desktop prefix IMHO is these days more of a problem with KDE apps (e.g. Konqueror, Kmail, etc) - they seem to go to some lengths to spell things with an initial 'K'. There are many mainstream GTK- or GNOME-based apps without an initial G (e.g. Nautilus, Evolution, Sylpheed, Pan) and with several that do, it stands for GNU anyway (e.g. the Gimp).

> ...the user should be able to (and usually can) run apps using either framework...

The thing people don't usually realise is that all running GNOME apps with GNOME, or KDE apps with KDE buys you is a consistent look-and-feel, and better desktop integration (and thanks to people like freedesktop.org, these differences are becoming less noticable all the time). But you can run them under fvwm with no "desktop" software at all if you prefer. They're X apps, and don't need anything more than a window manager to run.

Re:Desktop-specific afiliation

[dspeyer]

It is certainly possible -- I frequently do -- to run Gnome apps in KDE or vice versa or both in another environment. While basic interoperability works, they do look different. Some people have aesthetic objections to one scrollbar being blue while all the others are neon pink. (Personally, I have aesthetic objections to neon pink scrollbars, but that's irrelevant.)

Maybe somebody's already doing this, but wouldn't it be nice to have themes that made the differences go away? I realize this is what Redhat tried with Bluecurve, but Bluecurve was ugly. Perhaps the most popular themes for each hould be duplicated on the other. Maybe with matching mozilla and xmms skins too....

Of course, dialogs would still be an issue. It would take some real re-working to make Gnome and KDE share dialogs. Might still be possible....

Maybe I'll even do it, if no one else already is.

Now if we can just do something about OpenOffice....

And about my abuse of elipses....

Re:Desktop-specific afiliation

[FooBarWidget]

"but Bluecurve was ugly."

And this is exactly the problem. Everybody has different aesthetic preferences! Personally I like BlueCurve. There's no way people can agree on one single theme that everybody likes.

Apply unified theme A to GNOME and KDE -> some people say it's ugly.
Apply unified theme B -> other people say it's ugly.
Apply unified theme C -> yet other people say it's ugly.
Etc. etc.

Re:Desktop-specific afiliation

[BenjyD]

Linux is being split into two separate incompatible groups.

And what's wrong with that? Are you going to go round to all the KDE developers' houses and beat them until they work on Gnome (or vice versa)? Free software == choice.

Re:Desktop-specific afiliation

[Rysc]

What regexp are you using? It looks like Perl, but under Perl's regex $ is end of string/line and ^ is beginning.

$app =~ /^g/i;
etc.

Re:Desktop-specific afiliation

[Ben Hutchings]

There are many mainstream GTK- or GNOME-based apps without an initial G (e.g. Nautilus, Evolution, Sylpheed, Pan) and with several that do, it stands for GNU anyway (e.g. the Gimp).

Gtk was created for and is named after GIMP.

Re:Desktop-specific afiliation

[dspeyer]

I think you misunderstood me.

I'm not suggesting we force any theme on anyone (how coud we?) I'm suggesting we make some things available.

themes.freshmeat.net has popularities and ratings for its themes. If we take the highest, and bridge them over, and then publicize it (at least in the theme homepage, maybe we can get the freshmeat people to link it atomatically), then people who like those themes can use them everywhere.

If we can write some good automatic translation tools, then anything anyone wants should be doable quickly. Soon, anyone who wants a unified desktop will be able to have it without sacrificing aesthetics in their viewpoint

Re:Desktop-specific afiliation

[cobar]

You could just sort the themes on Freshmeat by popularity and contact the theme authors to request they do a port to the other toolkit, but a conversion tool would be even better.

What would be cool is if QT and GTK shipped with a theme that matched the other's default theme, so that way you could just flip a switch and have your QT apps match your Gnome desktop or vice versa.

Re:Other gnome stuff.

[Anonymous Coward]

a decent file dialog

This one I've never understood. Ximian came up with a decent (IMHO) file dialog for the GTK libraries they shipped with their version of gnome 1.2, and all their stuff was GPL'd. But here I am with a stock installation of Gnome 2.2, and there that stupid original file selector is...the only improvement in 2.x seems to be that when saving files, it no longer forgets the file name when you change to a different directory. Oh, they put some icons on the OK and Cancel buttons too.

Security Question.

[Anonymous Coward]

TheLastUser,

In your experience, can you comment on whether the Microsoft excel-format security flaws and macro virus exploits affect Gnumeric in any way? I am verry curious as to how Gnumeric implements an unstable Microsoft format without rendering some sort of security risk to a local user exploit.

Thankyou.

Re:Security Question.

[aurelien]

Format, I don't know. But VB is not supported as a scripting language, so Gnumeric is imune to macro virus exploits.

Re:Security Question.

[Jody Goldberg]

There are two somewhat related issues to contend with.

1) The file formats are semi documented. We have rough ideas of what OLE2, BIFF[5-8], and escher look like. There are however, lots of abiguities and question marks. As a result we have lots and lots of validation on what get imported. OOo does the same, which is why we can frequently crash MS excel when adding something new to the xls exporter, but still be able to read each others output.

2) The format for VBA is undocumented a far as I know. OOo has a few guesses in place and I've started doing some research on it, but neither of us can even read the vba enough to worry about running macro viri.

3) what scripting capabilities we do have (eg in python, perl, or guile) are strictly sandboxed. We are definitely tending to err on the side of caution rather than functionality.

Re:Security Question.

[sketerpot]

With the Python scripting support, what version of Python do you use? 2.2.3 is stable and well supported, 2.3b2 is faster and has features that make me very happy---but you can't have true sandboxing with anything later than 2.1, which is a bummer.

Re:Security Question.

[follower-fillet]

> but you can't have true sandboxing with anything later than 2.1
If I understand correctly, the reason post-2.1 doesn't have "true" sandboxing, is that the sandbox in previous versions actually didn't provide it either, just a false sense of security.

The decision was made to remove it entirely rather than continue to provide a dangerously flawed implementation. You can probably find exploits in the comp.lang.comp newsgroup.

Re:Security Question.

[jcast]

3) what scripting capabilities we do have (eg in python, perl, or guile) are strictly sandboxed. We are definitely tending to err on the side of caution rather than functionality.

Could you point me to some technical docs on how this works? I'm very curious about such things. Thanks.

Re:-21, flamebaitse.cx

[Anonymous Coward]

I agree, Linux security is appalling, we tried having Linux servers at work a few months ago, and every single one got hacked within a few days. "Secure" indeed!

Meanwhile we have never had to patch our IIS or MS SQL servers once since we've had them. They have remained very secure and reliable. "Linux is more secure" is complete and total bullshit.

Re:-21, flamebaitse.cx

[Anonymous Coward]

>2 GPL is anti-capitalist
this is bad how?

>3 Its hard to use
Depends on the end user. My mother has problems with Windows and that is classed as "easy to use"

>4 Its unstable
The kernel isnt, the X applications are.

>11 RMS is a communist arsehole
What has RMS done thats so bad you have to call im an arsehole? Just ignore him if bugs you.

>15 Its not from microsoft
you put this in the wrong list, that should go in the "Why Linux will survive list"
>17 Anyone and their 14 year old brother can add (buggy) code
Why does this bother you? the patch wont get accepted if obvious bugs are in it. And besides Windows has probably more bugs.

--voor

Gnumeric's best ideas are getting hard to clone!

[ratfynk]

It is a good thing that the open-source gcc is not a good thing to develop new ideas with for pirate-ware style software firms. Otherwise someone might succeed in patenting cloned ideas like a spread sheet, word processor, htm, internet commerce software interfaces.... digital communication firmware, encryption ssl, etc.

Hurray finally the folks who cloned visi-calc and 1-2-3 are starting to feel the heat. Bring on 2.6 and 64 bit AMD chips the fight is on!

Looks great, why not for Windows too?

[MadCow42]

It looks like a great replacement for Excel... why not make it buildable on Windows too?

I know that half the point of creating great desktop apps for Linux is to encourage the use of Linux on the desktop, but it also limits the usage (and therefor usage and availability of developer support too) of the product.

These days, there's almost no technical limitation to writing code that can be compiled on multiple platforms. Usually the limitation is the UI toolkit (gee, like Gnome?), but there are many cross-platform ones available too (like Tcl/Tk, etc.)

MadCow.

Re:Looks great, why not for Windows too?

[Jody Goldberg]

We're working on it. There is only one remaining technical requirement that is not available under win32, and the new egg menu/toolbar code will fill that slot nicely. Having a win32 build is a high priority. It may not make it in time for 1.2.0 next month but it will go in before the end of the year.

Re:Looks great, why not for Windows too?

[the-matt-mobile]

It looks like a great replacement for Excel... why not make it buildable on Windows too?

I'm glad someone asked this because I was just reading this article wondering the same thing. When I boot into my Linux partition, I'll occasionally try something that's Linux only like Gnumeric or Gnucash, but I find that it's too time consuming to learn the ins-and-out of all the Linux only incarnations of programs when I'm still primarily a Windows user. Programs like Gaim, GIMP, Dia, Mozilla, Apache, and OpenOffice are just more appealing programs to me because I can take their functionality back and forth between Windows and Linux. I don't know if there are many in this crowd that feel the same way about it, but from my standpoint if it runs on only one platform (Linux), then I'm no better off using it than I am the equivalent Windows-only program. I think a Windows version would make a big difference for some people.

Re:Looks great, why not for Windows too?

[2TecTom]

That's just so darn perversely sweet. I mean, I know it's supporting the Windows platform, but on the other hand, I can offer a practical alternative to limited income clients ~students mostly. This also greatly increases the chances of non-technical types adopting an open architecture.

Personally, I really appreciate the depth of features. I'd like to point out that an effective development enviroment is a core necessity. Perhaps not to professionals, but to the non-technical, working with known objects is often as far as they care to go into code. Indeed, VBA provides a very deep toolchest and often serves as the introduction to programming. It's great to see this done so well. Good hack, no doubtaboutit.

On this note, I curious though, what about interoperability? I've seen a lot of effort aimed at the internal functionality, but what about the external? I guess what I'm wondering is can users perform DDE and OLE type transactions?

Re:Looks great, why not for Windows too?

[Anonymous Coward]

Why don't YOU make it buildable on Windows?

Re:Looks great, why not for Windows too?

[MadCow42]

Because I'm not a cruel man, and I'd hate to abuse the current volunteers in that way. :)

Work with the OpenOffice team?

[Fragmented_Datagram]

I really wish there was more consolidation in the OSS world. It would be nice if the Gnumeric developers could spend their time making OpenOffice calc even better. Gnumeric may be good, but OpenOffice will be what the vast majority uses in the future...

Re:Work with the OpenOffice team?

[Anonymous Coward]

Wow. You can see into the future. You're amazing. Oh wait, no you can't. Open Office will not be what most people use in the long run. Odds are the MS Office will be dominate for another 10 years at least.

If they are knocked off the top it will be because the desktop wars have been won by either Gnome or KDE. Whichever wins will by default have their Office Suite become the leaded, just as MS is now because of their dominance.

OO is a stepping stone. Nothing more.

Re:Work with the OpenOffice team?

[Anonymous Coward]

The problem is, I, like the Gnumeric developers I'm sure, disagree with you. I don't like OO. Why can't Sun just work on Gnumeric :)

Re:Work with the OpenOffice team?

[Stinking Pig]

Could just as well turn the argument on its head though. OOo is good, but its spreadsheet component isn't as good as Gnumeric.

Re:For all the hatred OSS has towards MS product..

[Anonymous Coward]

Ummm, fucktard. Microsoft did not invent the spreadsheet. MS Office has become the standard for corporations all over the globe. While it's all fine and dandy to have brand new users use open source applications as their first computer experience, the real draw is taking market share from MS Office. Now go back under your fucking bridge and resume eating shit.

5 years and autofill still doesn't work right

[coaxial]

autofill doesn't understand how to fill two dimensions. autofill doesn't replace entries. Give me manual fill-right and fill-down any day.

Also dragging cells doesn't always figure out how to change the functions correctly.

excel is much better in these respects.

Re:5 years and autofill still doesn't work right

[Jody Goldberg]

File a few bugs with details of what you'd like to see added. Autofill in general is a vast swath of heuristics. Give us some clear specs and your pet peeves will be added.

I wouldn't turn down a patch either (hint hint).

Re:For all the hatred OSS has towards MS product..

[Anonymous Coward]

Since that time the project has grown to more than 300,000 lines and now supports all 325 worksheet functions in MS Excel, plus almost 100 more.

'Several' lines of code still in use...

[grimani]

...for example, like

int main(int argc, char** argv)
{

and even

}

at the end??

Not sure whether the choice of 'several' was just incompetence or failed sarcasm. Either way, it sure made my day :).

Re:'Several' lines of code still in use...

[Anonymous Coward]

Failed sarcasm? It was humour, damnit, and you picked up on it albeit assuming it was unintentional.

Gnumeric doesn't support Open/StarOffice format

[squashed]

Gnumeric is significantly faster for certain types of spreadsheet applications. I'd be happier if I could pick and choose among Open/StarOffice and Gnumeric. The best tool for the best job.

However, while they both support all sorts of Windows formats and predecessor Linux formats (OLEO, e.g.), they don't support each other's file format!

Creating Charts

[KidSock]

What's a good way to create charts (simple scatter plots) from data in a CSV file? Gnumeric just runs the CPU up to infinity when I try to create a chart with 10000 points. Any tips? Right now I have to use Excel :(

Re:Creating Charts

[Anonymous Coward]

Unless you're printing at 1200dpi, or rusing a printer roll, you won't *need* 10,000 points. reate a summary table.

If you feel you must, then there are plenty apps. Try using Perl to munge the data in, then print out as a postscript file.

Lastly, I doubt whether Excel is actually printing 10,000 points.

Re:Creating Charts

[KidSock]

I don't care about printing. I just want to look at the plot for 15 seconds before deleting it. I would do the same thing about once a day. Fact is Excel doesn't plot it the way I want either (but it does plot it without any problems which is better than nothing).

Re:Creating Charts

[kluro]

I would write a simple gnuplot script. Fast and easy to use.

Re:Creating Charts

[Bitsy Boffin]

gnuplot ?

Underwhelmed with Gnumeric's speed

[Megasphaera Elsdenii]

Feel free to mod me down, but working with
really large spreadsheets in Gnumeric is
a pain; it's way too slow. Reading in a tab-delimited
file with 12 columns and 40,000 rows takes minutes
(this is microarray data). I have compared
Kspread, Gnumeric, StarOffice, OpenOffice, even
Siag (scheme in a grid). They are all substantially
slower in than MS Excel ... For this kind of
work, I'm afraid I really see myself forced to
work with Excel (which, incidentally, runs
fine in Crossover Office; this is what I use on a
daily basis, because 1) Windows as a platform for
my kind of work is a joke and 2) I despise Microsoft)

In other words, if I had the time to do work on
Gnumeric, I would be only too happy to start working
on its speed when dealing with huge spreadsheets ...

Re:Underwhelmed with Gnumeric's speed

[Anonymous Coward]

Quite seriously, if you're the one with the tools to profile applications like this, might I recommend that you pull open gnumeric source and fix whatever is slow. After all, open source is about the freedom to do more than simply complain to the vendor.

Looks like there might be some plans.

[Chuck Chunder]

If you can infer from this comment from Jody.

Re:-21, flamebaitse.cx

[Anonymous Coward]

YHBT. YAG. FOAD.

Obligatory SCO joke

[cpn2000]

Yes ... but how long before SCO sues.

Re:Obligatory SCO joke

[kenstcyr]

Oh shit! Did they buy VisiCalc?

Re:Obligatory SCO joke

[mrjb]

a few days ago I was contemplating cutting of 4 fingers on each hand so I could count in binary.
Well leaving 2 fingers would actually get you ternary: 0,1,2,10,11,12, ... But I suspect it would never work. How would you cut the fingers of your second hand?

Re:Obligatory SCO joke

[Anonymous Coward]

But I suspect it would never work. How would you cut the fingers of your second hand?

I suspect that you could come up with some sort of foot-operated guillotine. I guess you'd still need help stopping the blood flow though.

any support for hexadecimal radix?

[Thinkit3]

Or at least mods for it? I don't like to use decimal.

Re:any support for hexadecimal radix?

[MrRage]

What? Do you have 16 fingers or something?

Re:any support for hexadecimal radix?

[Anonymous Coward]

Well, 14, technically.

YOU FAIL IT

[Anonymous Coward]

HA HA HA YEAH AND MAYBE THEY USE INCLUDE STDIO AND HA HA HA

no maybe they're not using { and } any more at all h ahaahahaha that'd be so fucking funny. my pants are filled with poo

Re:YOU FAIL IT

[Anonymous Coward]

Hey, at least "main()" was declared properly. The number of idiots who declare it void() on Slashdot is breathtaking.

I blame Python. Generations of idiots think they can "program" because they've written the odd Python script. Most of them wouldn't know a Quicksort from a hole-in-the-head.

Isn't it lame!

[Anonymous Coward]

That is the lamest excuse I have ever heard. What am I gonna do, recompile gnumeric for everyone in the office? Isn't the point of Linux that it's supposed to be BETTER than windows?

Jesus. Isn't it lame that the Space Shuttle uses thermal heat tiles instead of antigravity when entering the atmophere after deorbiting? What was NASA thinking!?!?! And on that note, isn't it lame that cars don't fly? Shouldn't they by now? And why are we still burning oil instead of using nuclear fusion? I swear, these idiots at NASA, Ford, and BP really need to get their asses in gear! Aren't we BETTER than this!?!?!?!

Welcome to the real world, idiot.

Re:TROLL STORE!

[Anonymous Coward]

Original source of photo for lemonparty

Hate the company, not the products (usually)

[Mr Z]

And Excel supports pretty much all the functions that Lotus 1-2-3 supported. Lotus 1-2-3 supported pretty much all the functions that Visicalc supported.

What did Isaac Newton say, again? I started over from scratch and ignored the work of those who studied these problems before I had? No. "If I've seen farther than others, it is because I have stood on the shoulders of giants."

Miguel has openly admired Microsoft's work in providing usable user-interfaces and applications that work well. He's also been critical of their excesses and lack of focus on security. Is it any surprise that Gnumeric (which aims to be able to import any Excel document) implements all of the Excel functions, but then extends them in its own way, adding nearly 100 of its own?

Personally, I don't particularly like Windows much because it doesn't work like I want to work. I'm accustomed to the Unix Way (or at least the Linux Way, though I did start with real UNIX in the form of AT&T SVR4, SunOS and Solaris). I really dislike Microsoft as a company, and anyone who thinks that removing choices and is a great way to make software easier to use. (Easier to learn, maybe, not not easier to use.) Hence, I don't run Windows at home, nor do I use MacOS X except via ssh. (My wife has a Mac in addition to a PC.)

--Joe

Re:Hate the company, not the products (usually)

[Rysc]

What did Isaac Newton say, again? I started over from scratch and ignored the work of those who studied these problems before I had? No. "If I've seen farther than others, it is because I have stood on the shoulders of giants."

I've heard it said that he trying to dismiss claims by a colleague of his, who was rather short, that without the colleague's work Newton would never have gotten anywhere. Thus, Newton was saying he stood, if anything, on the shoulders of giants, and not of this short man.

This is what I have heard.

Personally, I don't particularly like Windows much because it doesn't work like I want to work. I'm accustomed to the Unix Way (or at least the Linux Way, though I did start with real UNIX in the form of AT&T SVR4, SunOS and Solaris). I really dislike Microsoft as a company, and anyone who thinks that removing choices and is a great way to make software easier to use. (Easier to learn, maybe, not not easier to use.) Hence, I don't run Windows at home, nor do I use MacOS X except via ssh. (My wife has a Mac in addition to a PC.)

So do you really dislike GNOME, too? Because you know, they're all for the less-choice-is-easier thing.

Re:Hate the company, not the products (usually)

[Mr Z]

Actually, I'm no big fan of GNOME either for a lot of things (though I do like GNOME better than KDE). I find gnome-terminal obnoxious -- it eats so much CPU as compared to xterm, and it gets in my way in a number of ways. It's also noticably buggy.

Personally, in the GUI choices/configurability department, I feel that all the options should be accessible somewhere, even if it's behind "Advanced" tabs. Mozilla's about:config is a great catch-all, although I'd like a little more documentation on some of those fields before I change any of them.

So, yes, recent GNOME is slightly frustrating in some areas. I still get by with it just fine. I still prefer olvwm, though not by enough that I'm bothered to try to make it work anymore.

--Joe

Re:For all the hatred OSS has towards MS product..

[CaptainFrito]

well, um, M$ imitated 1-2-3 et al, then created file structures that locked everyone else's intellectual property (the contents of the spreadsheet in this case) into their own impenetrable file system theat ensures that you have to pay their toll just to share your work with someone else. One of the best features of Gnumeric, which I use often, is that the files it creates have a published scheme.

Still looking for decent charting app

[vrmlguy]

For the last couple of weeks, I've been looking for a charting application that's the equal of MS Excel. In particular, I have trending data where the X axis has to be dates, and I want to create .PNG images for use in a web-based application. So far, everything sucks. I've been reduced to trying to add date formating to plotutils, which isn't too easy.

Gnuplot seems pretty good, but isn't a GNU app (as I understand it, it semi-predates GNU) or much of an open source app. So GNOME feels that they can't use it and I don't want to use it for philosophical reasons.

Everything else, as I've said, sucks. Guppi looks interesting , though. I can't seem to find out if there's any way to use it from an Apache server-side app. Anyone else know?

Re:Still looking for decent charting app

[Jody Goldberg]

Please contact me (jody@gnome.org)

We've got a new charting engine in gnumeric now that will be released as a standalone library after gnumeric-1.2.x. Its goals are similar to gnumeric but on the charting side. Which means that it aims to have a superset of MS Excel's functionality. Thankfully that is alot easier for charts. The framework still needs a few extensions before I'm ready to split it out, but its already pretty capable. Adding a png or svg exporter would be fairly simple.

Re:Still looking for decent charting app

[damiam]

guppi is deprecated and unmaintained. The next Gnumeric will have a completely redone graphing framework.

Re:Still looking for decent charting app

[dodobh]

If you can script something out, Perl has this excellent module GD::Graph which will let you do that.

Re:Still looking for decent charting app

[Anonymous Coward]

You need chart director. I use it for various web-based charting, and it's much faster than ms charts.

Re:Still looking for decent charting app

[salesgeek]

If you would like a way to make instant raving fans, let's make the charting routines produce graphics that were made in 2003, not 1992 like Excel. Business communicators would be all over Gnumeric if:

* Ran on windows AND:
* Produced charts and graphs with true 3d, texture fills transparent mac aqua style bars/lines
* a few "themes" for graphs to make the non-designers (like me) look good
* allowed for a little display animation (i.e. growing bars, transitions between number sets (ie last year transform to this year)
* export to svg, gif and/or swf for animated charts.
* Support for WMF and EMF is critical for those who have to use PowerPoint/Word

Re:Still looking for decent charting app

[vrmlguy]

I had looked at GD and GD:Chart, but had somehow missed GD:Graph. It looks interesting. Thanks for the pointer.

Re:Comparing linux software to windows

[Anonymous Coward]

Actually if you check Slashdot's stats page, you'll find Windows is now in the minority. The breakdown is roughly 43% IE on Windows, 6% other browsers on Windows (including Mozilla and Opera), 29% Mac (X or 8/9), 18% Linux, and 4% "Other".

Windows is still the largest single group, but it's no longer in the majority. The big surprise is the Mac.

+1 Insightful.

[Anonymous Coward]

This comment should be "insightful" rather than "funny." When Gnumeric was created it was slapped together in a matter of days, mostly because of competition with KDE. An actual comment to the gnome mailing list was something along the lines of "yeah, we can hack up a spreadsheet in a matter of days" or "we hacked up the spreadsheet in just a few days." And the thing was, they were _proud_ of this fact. Scary. Glad I don't use GNOME.

Re:For all the hatred OSS has towards MS product..

[Anonymous Coward]

"What ever happened to original, non-legacy OSS programming?"

It never existed. All the OSS stuff is retro.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Text import limitations in 1.0.x

[Jody Goldberg]

We know.
Things are much nicer in 1.1.x

1) Better performance
2) A decent format selector
3) Configurable encoding and locale

#define is bad practice

[Beatlebum]

you should use const

System Requirements

[Jody Goldberg]

I don't really know. Up until last year my desktop was a Pentium 100 with 48Meg and things ran smoothly . I wouldn't want to build on that box, but as long as things don't get too big it should perform reasonably.

Re:System Requirements

[wmspringer]

Windows or Linux?

Re:System Requirements

[damiam]

Linux only (at the moment).

Our web pages need love

[Jody Goldberg]

1) Yes a couple of the routines are still subsets, but they tend to be corner cases (eg CELL). We'll need to finish them off before 2.0.

2) The web pages need work. I need to regenerate the function docs based on current CVS and setup some links from the status page to the docs.

Anyone interested in helping out ?

It could be worse - Re:All well and good, but...

[bazik]

$ emerge -p gnumeric

These are the packages that I would merge, in order:

Calculating dependencies ...done!
[ebuild N ] app-text/opensp-1.5-r1
[ebuild N ] app-text/sgml-common-0.6.3-r4
[ebuild N ] app-text/openjade-1.3.2-r1
[ebuild N ] dev-libs/libxslt-1.0.30-r1
[ebuild N ] app-text/docbook-dsssl-stylesheets-1.77-r2
[ebuil d N ] app-text/docbook-xsl-stylesheets-1.60.1
[ebuild&n bsp; N ] app-text/docbook-xml-dtd-4.1.2-r3
[ebuild N ] dev-util/gtk-doc-1.1
[ebuild N ] dev-libs/libole2-0.2.4-r1
[ebuild N ] dev-libs/libxml-1.8.17-r2
[ebuild N ] gnome-base/gnome-print-0.37
[ebuild N ] gnome-base/libglade-0.17-r6
[ebuild N ] gnome-base/gnome-common-1.2.4-r3
[ebuild N ] dev-util/indent-2.2.9
[ebuild N ] gnome-base/oaf-0.6.10
[ebuild N ] gnome-base/gconf-1.0.8-r5
[ebuild N ] gnome-base/gnome-mime-data-2.2.1
[ebuild N ] gnome-base/gnome-vfs-1.0.5-r3
[ebuild N ] gnome-extra/gal-0.24
[ebuild N ] net-libs/linc-1.0.3
[ebuild N ] gnome-base/ORBit2-2.6.2
[ebuild N ] gnome-base/gconf-2.2.1
[ebuild N ] gnome-base/bonobo-activation-2.2.2
[ebuild N ] gnome-base/libbonobo-2.2.3
[ebuild N ] app-text/docbook-xml-dtd-4.2
[ebuild N ] dev-python/PyXML-0.8.2
[ebuild N ] gnome-base/libglade-2.0.1
[ebuild N ] gnome-base/libgnomecanvas-2.2.1
[ebuild N ] net-nds/portmap-5b-r7
[ebuild N ] app-admin/fam-oss-2.6.10
[ebuild N ] gnome-base/gnome-vfs-2.2.5
[ebuild N ] gnome-base/libgnome-2.2.2
[ebuild N ] gnome-base/libbonoboui-2.2.2
[ebuild N ] gnome-base/libgnomeui-2.2.1
[ebuild N ] x11-themes/gnome-icon-theme-1.0.5
[ebuild N ] x11-themes/gtk-engines-metal-2.2.0
[ebuild N ] x11-themes/gtk-engines-thinice-2.0.2-r1
[ebuild&n bsp; N ] x11-themes/gtk-engines-redmond95-2.2.0
[ebuild&nb sp; N ] x11-themes/gtk-engines-pixbuf-2.2.0
[ebuild N ] x11-themes/gnome-themes-2.2.2
[ebuild N ] gnome-base/gail-1.2.2
[ebuild N ] gnome-base/eel-2.2.4
[ebuild N ] gnome-extra/libgsf-1.8.1
[ebuild N ] gnome-base/librsvg-2.2.5
[ebuild N ] media-gfx/eog-2.2.2
[ebuild U ] media-libs/freetype-1.3.1-r3 [2.1.4]
[ebuild N ] app-office/gnumeric-1.0.13-r1

$

I would kill myself now if my package management system couldnt handle dependencies :)

Re:Comparing linux software to windows

[Anonymous Coward]

mind to tell where the stats page is? I couldn't find it.

Re:Gnumeric _does_ support Open/StarOffice format

[Jody Goldberg]

1.1.x has an importer for sxc documents.
It could be improved, but the heavy lifting is in place and the rest just requires some attention to detail. An exporter will be added eventually.

I'm tempted to write a .gnumeric exporter for OO one day, but don't see much use in it given that we can read their native format and their xls.

Re:What does the F1rst post shirt say?

[Anonymous Coward]

Click the "larger images" button

Text import in 1.0.x was slow

[Jody Goldberg]

Please don't judge all of gnumeric based on the text import in 1.0.x. There have been lots of performance improvements and enhancements there in the development series. The core of gnumeric is easily capable of handling that magnitude of data. Try 1.2.x when it comes out next month (or even 1.1.x if you want to help beta things).

MS Excel is still somewhat faster mainly due to its memory foot print. It was written back in the day and bit bashes things all over the place. Gnumeric pays a penalty for using 32bit addresses rather than bit bashing 18.

If you have something that performs badly please _tell_ us. Our goal is to produce the best damn spreadsheet around. This is still version 1.1, 2.0 (extend) and 3.0 (extinguish) aren't due for a while yet.

Re:Text import in 1.0.x was slow

[johannesg]

Our goal is to produce the best damn spreadsheet around. This is still version 1.1, 2.0 (extend) and 3.0 (extinguish) aren't due for a while yet.

I commend you on your sense of ambition.

While I'm posting anyway, is there a method in Gnumeric to change a spreadsheet under control of another program in real time? I have a large body of data that is being gathered continuously and that my users like to look at in a spreadsheet while their process is running - does Gnumeric support such a thing? (I looked, but couldn't find it on the "features" page)

Unrelated to the previous question, does Gnumeric support timestamps with a resolution of less than a second?

Re:Text import in 1.0.x was slow

[Jody Goldberg]

1) depends what you mean by 'change' (I'm feeling clintonian :-). 1.1.x has hooks to support real time data feeds, and a sample implementation. That would allow some cells in a worksheet to depend on external content that updates. There is also a CORBA interface with a fair amount of bit rot.

2) We suppot MS Excel style format specifications so you can format a cell as mm:ss.000. Is that what you're looking for ?

Re:Text import in 1.0.x was slow

[johannesg]

1) That's the definition of "change" I was looking for ;-)

2) And again, yes that's the one I want. Thanks for the answers. The company I work for has become really open-minded towards Open Source lately (Open Office on every in-house desktop, Linux projects at customer sites, etc.) and a good spreadsheet (with these specific capabilities) is just the thing we need to solve a few problems.

Extending Gnumeric

[Jody Goldberg]

We tend to split extension into 2 areas

1) writing functions. Which is supported and documented in python, perl, and guile (and of course compiled languages)

2) scripting. Which is currently unfinished and intentionally mostly undocumented. There are some experimental bindings for python, but we have not had the time to select a solid enough api that we could commit to it. Gnumeric tries to under promise features, and I don't want to whip out some half baked api. The 1.3 development cycle will target scripting and we'll likely wrap the selected api in python, perl and corba initially.

We could use some help on this.

Other Spreadsheet Apps -- Terminal, even?

[weston]

Is anyone aware of any spreadsheet apps that will run in the terminal?

Other more off-the-radar spreadsheeting projects?

Re:Other Spreadsheet Apps -- Terminal, even?

[foobrain]

Yes, sc.

You can also use that example that came with TurboC under dosemu ":P

Re:Other Spreadsheet Apps -- Terminal, even?

[Argon]

GNU oleo comes to mind. There's also a program called sc.

Re:Other Spreadsheet Apps -- Terminal, even?

[eurostar]

Yes, Applixware is being used for this on
Linux, Windows, AIX, Sun Solaris, IRIX, HPUX, DEC, ....
scince 1983 and still going strong.

A new version is being released this year,
*not* the GTK version but the classical one.

It features new RealTime engines for Reuters RMDS and Tibco Rendezvous, these are for financial data feeds on the trade floor.

Re:Other Spreadsheet Apps -- Terminal, even?

[weston]

Are there downloads available or packages available for sale? I can't seem to find so much as a mention of such products on their site....

Re:Other Spreadsheet Apps -- Terminal, even?

[eurostar]

Applix went down the CRM business road, Applixware was inherited by Vistasource, who did a GTK variant, and a application server, but it wasn't a real upgrade path for most Applixware users. The "classical" Applixware that I am talking about has it's own widget set, and runs on legacy or low ressource hardware. ex. 11mb to start, and 2-3 Mb per app (TT, Spreadsheet etc) compare that to OpenOffice ressource requirements ! Check http://www.vistasource.com, or send a mail to sales@vistasource.com and ask them about the 4.4.x versions. The last version for Linux was version 4.4.2.

Gnumeric and non-English Excel

[Xolotl]

How well does Gnumeric handle xls files from non-English versions of Excel?

In particular, the formulae in non-English versions of Excel are saved into the xls files using their non-English names - can Gnumeric cope with that? (This is totally brain dead behaviour, IMHO, - not only does it mean that an English Excel can't understand non-English files, but if the function name has a non-Latin 1 character in it and you don't have that font, then even if you have the right language version of Excel you still can't edit the formula, only run it! This kills sharing Excel spreadsheets internationally. Why, oh why didn't they use numeric codes in the file and translate?). [Disclaimer: I've seen this for Excel = v.97, haven't looked at newer versions.]

As a side question, how does Gnumeric save formulae in its own-format files?

I originally tried Gnumeric a long time ago, in v. 0.something, at the time it didn't have the functionality I needed. I shall certainly try it again. Thanks for all the hard work!

Re:Gnumeric and non-English Excel

[Jody Goldberg]

We now use utf8 internally and pango to display content. I did a fair amount of testing last year importing various asian languages, and have recently started testing hebrew to validate the R to L support. So things should be on pretty solid ground for 1.2.

Our core file format is utf8 encoded xml, so its not really an issue.

Idiot

[Anonymous Coward]

Did you even read about it. It does have a new graphing engine, it's not gnuplot.

Re:Hypothetical MS question.

[Josh Booth]

But the irony is that if Microcrap didn't sell IBM Drek Operating System, Apple would have taken over, selling very proprietary systems, doing the same stuff Gates does, except also having a monopoly on hardware. OS X may never have went BSD and people would still be reverse engineering every change Apple made to their BIOS. Sure, smaller outfits would still exist with other computers, but they would probably not have been popular because they weren't Mac compatible. You would have the exact same shituation, but with one company controlling the hardware AND software. Of course, I don't have a beowulf cluster of universe simulating computers telling me what would have happened if Mr. True Believer Anonymous Coward had shot Bill, but I'm not sure the OSS movement would be less screwed.

Re:Comparing linux software to windows

[Anonymous Coward]

Check C:\WINNT\System32\Logs\W3svc

Pretty easy really.

99% perfect

[b17bmbr]

it is truly an awesome SS, except for one little thing. you can't formt the cells to have vertical text. the dialog box says unfinished. which says alot about OSS because ni commercial app would ever have that, but it still remains undone. i wish that one little thing would get done. if i was good enough i'd work on it, but i ain't. could some one please!!

Re:For all the hatred OSS has towards MS product..

[Josh Booth]

Since that time the project has grown to more than 300,000 lines and now supports all 325 worksheet functions in MS Excel, plus almost 100 more.

That's standard GNU software for you; the three options you actually use and the 29,854 you don't that were included because some programmer decided that "Hey wouldn't it be cool if my program could read some obscure file in some bastardized format, invented by some yak farmer in Tibet?"

Re:It could be worse - Re:All well and good, but..

[Anonymous Coward]

Of course, that's gnumeric 1, which kind of sucks ass. Looks to me like the ebuild's dependencies suck ass, too, since it's a GTK+ 1 app and half the installed libraries are GTK+ 2

Re:For all the hatred OSS has towards MS product..

[JimmytheGeek]

Let's see - so the OSS app gets criticized for copying Excel, which itself copied lotus 1-2-3. THen you criticize it for exceeding excel. Fucktard.

Re:Yes i would

[Anonymous Coward]

If you're happy to commit genocide, I think you can have the title of "Kind of oppression."

Re:Gnumeric _does_ support Open/StarOffice format

[Anonymous Coward]

Pull a Microsoft. Make your format so cryptic and complicated that no one else can read it. Then just support everyone else's format. 3. Profit!

Gnumeric Kicks Tail

[rinkjustice]

I've tried alot of spreadsheet programs on both sides of the fence. Whilst taking a MS Excel course a few months ago, I wanted to use a free alternative and not a warez copy of Microsoft Office for obvious moral reasons. And I sure as hell wasn't going to pay for a copy! I've tried 602Tab, an Excel clone and part of the impressive 602 PC Suite for Windows, KDE's spreadsheet program (the word "kludgy" comes to mind) and OpenOffice which was so damn sluggish I gave it the ole make uninstall; make clean routine after 5 nerve-fraying minutes.

If there's a better (and free) spreadsheet program out there, I haven't used it.

Re:For all the hatred OSS has towards MS product..

[Anonymous Coward]

Haven't you noticed that much of the OSS community spends its time copying Microsoft?

I realized long ago, a lot of OSS isn't about "freedom", it's about "free". As in beer.

The rhetoric is just a cover.

Shouldn't you be...

[Anonymous Coward]

waddling around outside on flabby thighs playing with lawn darts and drinking your disgusting swill beer? Today is the birthday of your country and a day of mourning for people sick of your fucking culture and shitty products.
Fucking americans are disgusting

-Troed

Re:Shouldn't you be...

[Anonymous Coward]

We played croquet. Fucker.

References and indirect lookups?

[mercuryresearch]

Anyone know what the current state of different reference systems (R1C1 and A1) are, as well as their use in the INDIRECT() statement? Last time I checked this was still broken, though there was some discussion about it being fixed in the CVS version.

Internally we use some pretty complex spreadsheets and a few break gnumeric, which is one reason excel is still around. (They sometimes break excel, too.)

I've moved quite a few of the smaller but still important items to gnumeric, and it's not let me down yet. And the XML file format is great -- in lieu of a real scripting API I've used PyXML and python to read the Gnumeric files and process things that way, and it was really pretty easy to do.

I tried it in 1999 and it sucked....

[Anonymous Coward]

And I thought it probably still sucked until this day! Gee, I ought to try out all them other "previously sucky" applications that come with my Linux Distro to see if they no longer suck.
How many other projects that we once tried out in their beginning stages that sucked are now great and useful tools today?
I think Khexedit turned out pretty well....
By the way, what ever happened to killustrator...I remember it changed names, but I also remember that it didn't have much functionality...has it surpassed the real Illustrator or has it died a slow and painful death?

Re:For all the hatred OSS has towards MS product..

[Anonymous Coward]

I've noticed this myself. <troll> OSS developers obviously have very little imagination </troll>

Sad != happy

[Anonymous Coward]

In other news, your mom turns 18.
See the story at 10:00

Gnumeric is what Unix office apps should be

[Argon]

Hi,

I don't use office applications a lot; but Gnumeric is one application that showed promise from the very start. I have used Excel occasionally and I was immediately at home with Gnumeric - even in the early days. The interface is snappy; the GUI intuitive and it could import Excel files without a problem. I've looked at many office applications on Unix and I can honestly say that Gnumeric tops all of them. Now if only I had a Power Point equivalent that is as good as Gnumeric ...

Re:Lines of code still in GNUmeric...

[Anonymous Coward]

YOU VIOLATE THE HOLY K&R WHICH CLEARLY STATES '{'s should be on a new line!!!

MOD PARENT UP!!!! IT AINT A TROLL!!!

[Anonymous Coward]

It includes an interesting question:
How many other projects that we once tried out in their beginning stages that sucked are now great and useful tools today?

Gnumeric doesn't do Pivot Table

[Taco Cowboy]

Contrary to what have been advertised, there _are_ something that Excel does, that Gnumeric doesn't.

For example, Gnumeric still doesn't do Pivot Table.

While I do understand that Pivot Table isn't really a big deal to many, there are times that functions such as Pivot Table comes _very_ handy.

Please, Gnumeric Developers, please put the Pivot Table on top of your "todo list".

Thank you !

Re:Gnumeric doesn't do Pivot Table

[Jody Goldberg]

There are certainly features of MS Excel that Gnumeric does not support yet. Pivot Tables and Conditional Formats are the main outstanding issues and are targeted along with scripting and accessibility as the top priorities for the 1.3 development cycle in the fall.

The 100% figure refered only to the worksheet function coverage. Sorry if that was not sufficiently clear.

-1, Irrelevant rant

[0x0d0a]

The random "GNOME sucks" rant was totally unrelated to the post that you were responding to.

CONGRATS JODY GOLDBERG

[bongobongo]

on the largest single-day karma take in slashdot history!!! ;x

3D spreadsheets

[ToadMan8]

This might be too late at night for me to post coherently, but I have been looking to ask some geeks about this lately and this is the perfect forum.

I wish spreadsheets worked with three dimensions. For example, say you have like 10 parts you are measuring the sizes of. The dimension of the part you measure makes up the columns (Y axis) and the 10 parts makes up the rows (X axis). Now what if you have five variations of parts? You have to make five different sheets, and none of the graphing and data analysis is easy to work with whatsoever. Now, if there was a Z axis to the spreadsheet.... then the variations would be deep... That would be spectacularly easy to manage data of this type with. Is there anything out there that does this? Make it! Heh. I'm going to bed. G'night folks (it's 3:44 AM my time).

Re:3D spreadsheets

[Anonymous Coward]

Dude, that's a relational database. :-)

Re:3D spreadsheets

[EarthVsMe]

Well, my dad used to use something called QubeCalc on MS-DOS, which he said was a 3D spreadsheet. Still, I don't think it does graphing or anything spectacular, seeing as the latest version is dated December 1989.

Re:3D spreadsheets

[WillAdams]

Been done before, but it's just not survived.

Look up the spreadsheets Javelin, Lotus Improv and Lighthouse Design's Quantrix. There's an on-going attempt to revive Improv / Quantrix for Mac OS X (they were originally NeXTstep programs), see http://www.materialarts.com/FlexiSheet/

The wikipedia has a good page on Lotus Improv:
http://www.wikipedia.org/wiki/Lotus_Impro v

William
(who once tried to inform the ignorance of an author of an encyclopedia article who was blissfully ignorant of 3D spreadsheets)

Slightly OT

[burns210]

"Five years ago, Miguel committed the first code for Gnumeric to CVS. In a testament to the quality of the code several lines are still in use."

Slightly offtopic(sorry!), but i wonder if this can be said of linux. What is the oldest code that has proven the test of time in the linux kernel and is their anything from the way back 1.x or 0.x days, or has it all been replaced(for better or worse)?

Re:Gnumeric _does_ support Open/StarOffice format

[burns210]

how about we (the open source office apps) do what we have been talking about, (i think OOo has started this) and move to a common, XML, buzzword based fileformat, so we wouldn't have to save in OOo or gnumeric or kword formats rather in the opentext format, openspread format, etc...

come on people, we complain that proprietary software should use common protocols and standards rather than closed, proprietary ones, but we are doing the same thing by not sharing fileformats across a compe standard!

Re:Comparing linux software to windows

[spaic]

It's compared to MS Excel simply because it's the best spreadsheat application around. Just as the Gimp is often compared to Adobe Photoshop (originally a mac application) and MS IIS to Apache.

Re:Comparing linux software to windows

[Anonymous Coward]

What do you mean? What does the slashdot stats page have to do with C:\WINNT\System32\Logs\W3svc?

I am not even using Windoze.

I MADE YOU LOOK!!!! NA NA NANA NA!!!

[Anonymous Coward]

:-p

Re:Lines of code still in GNUmeric...

[Doomrat]

Personally, I actually DO put the braces on new lines, but not in this case as I wanted to make the code a bit longer to aid the lame joke.

Re:Lines of code still in GNUmeric...

[Anonymous Coward]

This is number 5... please stop these stupid geeky comments :(

Excel

[Anonymous Coward]

Too bad they are copying the most crappy program in the MS Officesuite.

It is missused by many as a database.
It would be better if they did develop an Access replacement. Fast db building, in a new way.
Easy to use, connecting to any OS db.
They do not have to make any special DB specific stuff.

A REPORT GENERATOR and a Formbuilder.. would be what is needed.. Excel can ALWAYS be replaced by a better DB, but it just has to be as easy to create.
Try implementing a DB in mysql and do some nice reports
in an hour. Then try the same thing in Access..
And it is not like Access is a new kid on the block.
I say OS can do it better..

Re:For all the hatred OSS has towards MS product..

[Anonymous Coward]

It never existed. All the OSS stuff is retro.

Tex?
Mosaic?

I use Gnumeric, but...

[ChaoticCoyote]

...not for serious data analysis. The graphing modules are weak, and it still tends to crash at in opportune moments. Try sorting a table containing forty columns, for instance, and see how the dialog box refuses to list all the column headings.

And Gnumeric's greatest weakness -- it is, for the most part, a direct clone of Excel. No real innovation, no new ideas, just an imitation that conveniently runs on my Linux systems. Convenience is nice, but I'd like to see some serious attempts at improving the spreadsheet model.

Re:I use Gnumeric, but...

[Jody Goldberg]

1) If you can crash it please _tell_ us s othat we can fix it. 1.0.x has been very stable.

2) The interface to guppi was weak in 1.0, the new engine in 1.1 is much better integrated.

3) I'm not clear what you mean for the sorting dialog. File a bug.

4) We have started moving past MS Excel in places. Improvements to dialogs, more functions. We first need to have a foundation that can cover the existing fetures before adding wild new ones or no one will be able to leave MS Office. People generally want to bring their data with them to new applications.

Picky Picky Picky Gnumeric Posts

[Linker3000]

To all those posting whingey, whiney comments:- a) If you don't like the software ask for your money back or better still, contribute some time and effort to help improve the software if you have the skills. L3K. PS: Try the same tactics with any M$ app!

Re:It could be worse - Re:All well and good, but..

[Anonymous Coward]

I remember trying to upgrade gnumeric back in the day when I was using RedHat. It was an experience that I don't think that I could ever forget. I think I spent days trying to satisfy all the dependencies....

Needless to say, I now run Debian.

Re:Lines of code still in GNUmeric...

[An Onerous Coward]

I've found some other lines that have survived: /*** Drunk. Fix later. ***/ /*** Something in here is causing segfaults ***/ /*** The next 300 lines of code "borrowed" from SCO Unix. They won't miss it. ***/

Quality of code

[peterpi]

"In a testament to the quality of the code several lines are still in use."

int main (int argc, char ** argv)

Gnumeric and Open Office

[CausticWindow]

The right direction for Gumeric, would be to merge it in as the spreadsheet in Open Office.

Gnot what you think

[braman]

The "gn" in most open source applications stands not for "GNOME", but for "GNU". GNU, in case you haven't heard, stands for "GNU's Not Unix!", one of the earliest and essential parts of the free (as in speech) software movement. This used to divide Gnome and KDE folk because KDE, early on, was not free. That is (to some degree, at least) no longer the case. HTH.

Re:Gnot what you think

[Anonymous Coward]

And you would be completely misinformed. The "G" at the start of applications written with the gnome-libs, etc. has _always_ meant GNOME rather than GNU.

This used to divide Gnome and KDE folk because KDE, early on, was not free.

Again, you simply fail to understand. _Miguel_ was the sole cause between the Gnome vs. KDE divide. KDE itself was _always_ free and GPL. Simply because it uses a widget toolkit which was less-than-free does _not_ make KDE itself less-than-free. It was entirely possible to recreate a free Qt (which did happen to some extent). Miguel convinced a large population that KDE was close to proprietary, or not free. The whole reason for starting GNOME was because supposidly, KDE was not free (Miguel's thoughts). Now that KDE is completely free (Miguel's "free") why do people continue on working on GNOME? Unified desktop my ass. This is about Miguel and Miguel's ego and business: Ximian. Nothing more.

-1, Offtopic

[Anonymous Coward]

Frankly, I never said "GNOME sucks." I simply said it's Miguel's little ego-boosting money-making parade. Which it is. Which just happens to be completely related to the original post, unlike your (and this) offtopic post. Grab a clue bat and hit yourself a good one, buddy.

Excel's plotting still better than gnumeric

[Anonymous Coward]

Excel may not be great at plotting, but it's not horrible, and it's decent for a spreadsheet. And I know people suggest using gnuplot or something, but that's not an acceptable solution when I'm manipulating a spreadsheet and want to see what my data look like. I want to highlight two columns, click chart, tell it I want an XY plot, and have it do it. I don't want to export it, write a script, etc.

I really like gnumeric, but its lack of plotting capabilities have prevented me from adopting it fulltime. Otherwise, I like its analysis tools and compatibility with MS, not to mention it's very light compared to either MS or OO.

Re:Comparing linux software to windows

[Anonymous Coward]

I am not even using Windoze.

Too bad, because Linux doesn't have that directory.

Re:-21, flamebaitse.cx

[Anonymous Coward]

One reason why you deffer from human race.

YOU'RE WAY TOO STUPID

Re:-21, flamebaitse.cx

[justsomebody]

I see the reason why you post as AC

ui designers needed

[10bt]

nice work for an oss project, but the ui still needs some polishing. i was looking at the screenshots and noticed that some things were obviously designed by engineers not ui designers or graphic artists (e.g., there is no padding around the text label in the "print preview" button). i guess this makes since most artists and designers are not hardcore programmers.

Re:ui designers needed

[Jody Goldberg]

screenshots are out of date. They correspond to the 1.0.x stable series rather than the friendly new 1.1.x versions.

However, we'd love to get more UI review.

Re:Gnumeric _does_ support Open/StarOffice format

[Ben Hutchings]

The trouble with this is, new features in the application will often require additional data to be stored in the files. An interchange format is good for transferring lowest common denominator data, but there's still room for application-specific formats. XML-based interchange formats could be extensible (it's what the X stands for!) to allow applications to add features without making their files unreadable. This will only work as long as the extensions aren't critical to correct interpretation of the file, e.g. macros in a spreadsheet that are written in a non-standard language.

what's that mean?

[Thinkit3]

Besides the number of body digits in hex.