What OSS Programs are Still Needed?

suso asks: "I was thinking yesterday about how much open source software is out there already. Most categories are filled, but I wonder about what pieces of software still need to be written. What programs would you like to see in OSS form that are currently not available?"

Comments

Voting software.

[Sierran]

Anything to put Diebold out of business, thank you. Auditable, open-source, tested. Please. Before the U.S. Midterms.

Re:Voting software.

[Tumbleweed]

Unfortunately, that would have to be a software AND hardware solution, to be anything remotely effective.

Re:Voting software.

[np_bernstein]

The Open Voting Consortium is doing a pretty good job. The program itself is written in python which is a good idea (despite being a perl guy) since there is always the risk that someone could keep additional code out of CVS for compile time, has paper verification w/ barcodes for quick recounts/auditing... all in all they're doing a pretty good job.

I understand that there are hardware requirements for security as well, but this provides a good way for businesses to still sell and add value to this software.

They are currently having a fund drive so if this seems like something important, consider giving them $10 (what they're asking for)

Re:Voting software.

[sampowers]

Yes, but the most important part of that is that the software is as fault-free as possible. You've read all the reports of problems with this election.

Design your software as if your hardware had no security, and design your hardware as if your software had no security. Unfortunately, Diebold seems to have this completely backwards.

Re:Voting software.

[Tumbleweed]

Actually both parts are equally important. 'Fault-free' software means nothing if the machine doesn't make a paper copy for a physical trail.

I've actually got a design in mind for this problem, but it might be too expensive - I'm not sure how much those machines cost.

Re:Voting software.

[Bachus9000]

When democracy is at stake, no price is too high. :)

Re:Voting software.

[Anonymous Coward]

Voting systems should be EAL7 certified. The open voting consortium's FAQ doesn't mention anything about evaluation or assurance at all.

Re:Voting software.

[np_bernstein]

I think they need to finish the software first, and then they can start looking at people to audit the software.

DRM

[sevinkey]

Need a reliable open source DRM solution that can be proven trustworthy so we can finally have pay content that's not platform dependent.

If you can build a linux box for $70 and call it a linksys router, then with a OSS DRM you should be able to create the equivalent of MCE2005 for $250.

Re:DRM

[Creepy Crawler]

With an open system, copying the unencrypted stream is always possible (via /proc/kcore even).

You need something secure from the driver up to support crap like Digital Restrction Management.

There's 2 defences that the media companies have. One is the law. Two is heavy watermarking of movie files. If you cant guarantee to prevent copying, put the peoples' name and computer hardware information inside it.

Re:DRM

[finkployd]

DRM is a fundamentally flawed concept. The only implementations of it that exist involve basically turning the concept of public key cryptography upside down. Giving someone a private key with which to decrypt data but then trying to prevent the user from ever coming in contact with that key is just silly.

The way it is SUPPOSED to work is the owner of the key is the only one who controls it and nobody else ever comes in contact with it. The way DRM is supposed to work is by having the content producers control what you can and cannot do with your own private key, and try to keep it from you.

This is why DRM (in its current form) is so easily (and constantly) broken and also why it will never work in an OSS implementation. Even strong proponants of DRM (you can find them at the Digital Identity World conference pontificating about how great DRM will be for the computer industry) admit that someone will be able to break it, they just want to keep everyone from breaking it. Of course once one person breaks it it and unencumbered media is released, then it is all over anyway.

Finkployd

Re:DRM

[BillyBlaze]

DRM relies on security by obscurity, either at the software or the hardware level. (Software that just won't do what users want, or hardware that withholds keys from the user.) In hardware, this is at least somewhat viable, because it's hard to get info from chips that don't want to give it. But for software, it's not viable, because any software can easily be modified. Open source solutions would be even more easily circumvented, because they can be even more easily modified.

I think the best way for open-source users to win the DRM game is the same way everybody else can win - don't play.

Re:DRM

[cgenman]

Unfortunately, you can't open-source DRM. At a fundamental level, you can't prevent the computer from having enough information to decrypt a file, without the user also having enough information to decrypt the file. The only way to do that would be to store the key in the decryption hardware in a write/decrypt only location. Even then you'd have to have special hardware to intercept the decryption key as it is transmitted to the machine and before it hits the processor, in order to deflect it to the proper area without the machine having any say. Even then you have man-in-the-middle attacks, which would require a layer of communication encryption that the processor doesn't have control over, etc.

In other words, you fundamentally cannot give the user full knowledge without relinquishing full control.

Re:DRM

[lithiumcloud]

/In other words, you fundamentally cannot give the user full knowledge without relinquishing full control./

Orwellian.

A 100% functional and universal audio mixer

[Beatbyte]

Having been on linux 100% for the last 3 years, I've never had an audio mixer work right (even on a sound blaster live! value) out of the box.

The channels are not labelled correctly, the fader doesn't work on most of them, the inputs are changed on the back of the card (i.e. from rear out to line in!).

The only reason this is so important to me is the 5.1 surround setup I have has no volume control. The volume is at whatever level the computer sends to it. I'm sure this is not the standard setup so it doesn't get much attention.

Re:A 100% functional and universal audio mixer

[Anonymous Coward]

Audio on Linux in general is a mess.

Trying to do audio recording on Linux is essentially futile.

Re:A 100% functional and universal audio mixer

[Noksagt]

ALSA in modern kernels really has improved things a lot over that time, though. In my boxes, I didn't have any headaches at all. If you have a well-known card, there is probably at least a work-around. Channels with the wrong PCM names are easily fixed. Sending a report on the exact make & model of your card and the problems you're having may improve the .asoundrc used for your card in the future.

Sound still isn't perfect, but it is a lot better than it used to be.

Re:A 100% functional and universal audio mixer

[dhakbar]

I highly doubt that whoever modded the parent as flamebait has ever tried to do any kind of audio work in Linux...

Re:A 100% functional and universal audio mixer

[Anonymous Coward]

alsamixer is about the only one that gets it right most of the time. gnome-alsamixer is OK, but, it messes up with my SB-MP3-USB. I've tried an awful lot of things, and had little success... many programs dont even let oyu specify the mixer device, and only support OSS.

Re:A 100% functional and universal audio mixer

[dalutong]

audio editing software. This includes programs like ProTools, FruityLoops, etc. While a little bit exists it all competes very weakly with the closed-source software.

Re:A 100% functional and universal audio mixer

[smeat]

Well let's see, I installed Mandrakelinux 10.1, then installed the DeCSS libraries frpm plf that are a requirement thanks to the MPAA, then I had full 5.1 surround sound on all DVD movies I have played. Gosh the amount of work involved is astronomical. No one would even be able to figure that out.

smeat!

Re:A 100% functional and universal audio mixer

[Anonymous Coward]

Watching a DVD does not qualify as "audio work", idiot.

Re:A 100% functional and universal audio mixer

[Noksagt]

I wouldn't disagree 5 years ago. ALSA is fantastic today, though.

There are also plenty of custom packages with Linux audio software bundled with extremely high-end, multi-channel cards.

Re:A 100% functional and universal audio mixer

[GigsVT]

Audacity is pretty good. Not great but still.

Re:A 100% functional and universal audio mixer

[WoodenRobot]

This is the one thing that stops me from using Linux 100% of the time. If only there was some replacement for FruityLoops and Ableton Live, I'd never have to touch XP again, which would be a god-send.

Re:A 100% functional and universal audio mixer

[dalutong]

yup. we're in the same boat. my laptop is 100% linux. but i have to work on windows machines to make music.

Database

[adam mcmaster]

I hate to say this, but a database frontend that's as good as MS Access would be nice - there are attempts at such applications currently in development, but nothing that's even close to usable.

Re:Database

[Anonymous Coward]

Oh, there are plenty of options out there, you just haven't found the one that gets it all 'right' for you yet. Try Eclipse with the Quanta or DB2/derby plugins. Or other DBMS's have their own tools, one will surely fit your own intuition.

Granted, though, there are a heck of a lot of OSS database frontends that are far from usable...

Re:Database

[mini me]

Rekall doesn't seem too bad. But I think the real question is, why would you want to use something like MS Access at all?

Re:Database

[Rysc]

There are two theoeries as to why Access is bad:

1) It's DB engine is primative and a joke making all databases created for it fundamentally inferior.

2) It dumbs down databases making DBAs annoyed that their boss' can say "My 12 year old sun threw together a database that works fine in 20 minutes! Why am I paying you?"

The first argument is the only one which is good. You do NOT want to use a technically inferior DB.

The second argument is NOT good. For some people, for some things, you really do not need a DBA. The fact that people CAN build databases without really understanding them is not fundamentally bad, it's an empowerment and fundamentally good... so long as it is understood (as, of course, it some times will not be) that a DB designed by an amature in a WYZIWYG DB app is not the same as a DB designed by a professional, just as some kids VB bitmap editor is not Photoshop.

Free software, in my view, is ultimately about empowerment: My ability to do more. Not necessarily without knowing more, but without spending more, and without being forced to do it someone else's way. To empower more people it can be necessary to allow for people to do more while knowing the same or less (see some aspects of the GNOME philosophy of late). That isn't bad, in fact it's definitively good.

So, just as a WYZIWYG html editor is not bad just because FrontPage blows, a GUI database designer is not bad just because Access blows.

Re:Database

[drig]

The Access interface is nice. I know the database underneath is awful, but the front-end is powerful and easy to use. The best use of it I've seen is to make reports based on data in your database. Sure, you can use Crystal or Hyperion or something like that, but Access is much easier for quick "show me the break-down of this set of data" queries.

Re:Database

[0racle]

The access database format is what drives MS MOM 2000. Its not that bad, it just wasn't meant for huge databases.

Yes I know that MS MOM says MS twice, would you have known what I was talking about if I had just said access drives MOM?

Re:Database

[megabyte405]

GnomeDB Mergeant ?

Re:Database

[Unordained]

The problem I've always found with letting users design their own databases (that's when you can even keep them from using Excel to do it) is that they don't understand *basic* cardinality issues. They never think about how many of "these" go to how many of "those" or how they want to avoid re-entering data, make sure they don't modify something they don't mean to, or constrain their numbers to actually line up the way they want them to (assuming they can -- it's amazing how hard it can be to explain that A + B != C in a particular situation, no matter how much you want it to be that way.)

WYSIWYG is fine and good (and even to be encouraged -- databases are not so sacred or complicated that the average joe *shouldn't* be able to build his own to suit his needs,) but could it please come with a good tutorial/wizard that helps the user think through the issues? What entities do you have? How are they related? What changes often? Tutorials and wizards to help them expand the database as their needs change would be good too -- help them move the data into a new layout without breaking anything (while it knows what's going on, it could even create backward-compatible views in some cases), etc. There are a lot of re-used ideas in most databases, related to keeping ordered lists, trees, graphs, non-overlapping date periods, etc. and those should be made easily accessible (and extensible?)

Whatever it is, it needs to be as easy to use as a spreadsheet plus a little bit -- move people away from spreadsheets as much as possible. Having a combined layout + data tool is fun, but it creates absolutely abysmal habits that only result in bad data. And then I have to clean up the mess.

Re:Database

[Anonymous Coward]

I recognize people by their sigs. Is that a bad thing?

Ditto!

Re:Database

[mandreiana]

See OpenOffice 1.9.60 Base

Re:Database

[jonadab]

> could it please come with a good tutorial/wizard that helps the user think
> through the issues?

People who use wizards don't think through issues. They click "Next" until
it's all done. Conversely, people who think through issues don't like wizards,
because they're grossly inefficient (in terms of the number of screens, number
of clicks, and so on) for what they accomplish.

As far as database GUIs, I think they're great for *most* databases. *Most*
of the world's databases consist of data that can readily be stored in a
single table, with a relative handful of columns (no more than perhaps 20)
and a single row for each record. Heck, fully *half* of the world's databases
are essentially a list of names with attached addresses and phone numbers and
stuff. You don't need a DBA for that, nor SQL, nor even a computer geek
really. The database in MS Works 4.x is good enough. You can sort by any
column with a single click, how convenient is that?

When you start needing multiple tables, or having multiple rows that logically
belong together (e.g., some for people you have multiple addresses and phone
numbers beyond there home/work ones that you have columns for), you start
going beyond the boundaries of what the non-geek user can handle doing with
a GUI, and at that point you want a relational database and a geek. If it's
only a couple of records pushing your boundaries that way, you deal with it
in a Notes column, no big deal, but when it starts being a significant number
of records, or if you just plain need multiple tables, it's time to get a
geek, someone with a mind trained to think in layers of abstraction and
organize data in more than two dimensions -- and he's probably going to want
a relational database.

Honestly, I don't think a wizard or anything like that can fundamentally
change this. You might push back the edge a little, but not very far.

Re:Database

[Anonymous Coward]

It's DB engine is primative

"Its".

Re:Database

[Rysc]

You are absolutely correct and I thank you.

CAD

[ikeleib]

Serious CAD software. There are a few projects that do some 2D drafting. That is not sufficient. A serious CAD package can not only serve as an end application, but the backend to many tools.

I have thought on more than on occasion of starting a foundation to get such an effort off the ground. I felt then as I do now that there are many places that would contribute serious money to the effort of an OSS CAD. Organizations spend serious money on CAD. Additionally, there is lots of out of work talent that would be willing to devote serious time to such a project if it were financially possible for them.

Re:CAD

[Anonymous Coward]

Mod him up, I know people that the only reason why they do not switch to linux is because the lack of high end CAD.

Re:CAD

[QuantumG]

what software does this on other platforms? Can you give us some names? In general you're not going to get specialty software written with the OSS model. The people who use the software must be the authors. Either by writing the code themselves or by hiring others.

Re:CAD

[maeka]

Amen!

And to agree with the AC above me - THE reason I have not switched to Linux is Autodesk Land Desktop - period.

A good, OS, 3D CAD program would be a godsend.
I do realize how many millions of man-hours must be incorporated in AutoCad 2005, much less Land/Architectural/Mechanical Desktop. It must be a high hill to climb, for any new project - not to mention all all the plug-ins/add-ons available for AutoCad...

I could be tempted to switch from Windows if there was something with only R14 or Microstation V7 capabilities.

Re:CAD

[Anonymous Coward]

I think you're confusing "OSS" with "runs on Linux".

Re:CAD

[Noksagt]

Open CASCADE is a 3D CAD. It started out as high end CAD software, didn't keep up, was bought, and turned open source. Please not it is made available under a special license that, while resembling the LGPL, has (AFAIK) not been approved or even submitted for OSS certification.

Re:CAD

[Noksagt]

CAD--Computer Aided Drafting, such as AutoCAD, SolidWorks, Mechanical Desktop, etc. is the sort of specialty software I think WOULD be OSS. (1)It is a technical product--a few of the engineers who would use it would also have the skills to code it. (2)It is fairly simple to steal components from rendering programs and other existing programs that are already F/OSS. (3)It could be extended for use by FEM programs that academics & others who have more time and freedom to develop F/OSS would absolutely love to have.

Re:CAD

[ikeleib]

Open CASCADE is an ingredient in a CAD system. From their web site: Open CASCADE Technology can be best applied in development of numerical simulation software including CAD/CAM/CAE, AEC and GIS, as well as PDM applications.

I would fully expect an OSS system to use Open CASCADE. As far as I know, there is no other geometric engine that is anywhere near as advanced in the OSS space.

Re:CAD

[Noksagt]

Salome is one such LGPL program that is built on Open CASCADE and other F/OSS projects. Looks quite promising.

Re:CAD

[maeka]

I think you're confusing "OSS" with "runs on Linux".

Any good OSS WILL run on Linux. If the source is there and the program is tasty - it will be ported.

Re:CAD

[dourk]

To big a project for too small an audience, I think. How many years would it take to build an open-source app equivelant to solidworks?

I've spent serious cash on cad/cam software, and I'll continue to spend it until a tool comes along that can help me design and program parts faster.

Which means I might try an OS cad program, maybe once a year, get disappointed after about 20 minutes, and delete it for another year.

It wouldn't have the loyal following of users that strive to make it a competetive product.

If the Gimp isn't "as good" as Photoshop, big deal. If OScad isn't at least as good as Solidworks, I'm losing money.

Re:CAD

[j1m 5n0w]

ME-10 and ME-30 are a good example of how to do 2-d drafting right, and Solid Designer is very nice for 3-d models. Solid modeling is definitely the way to go. Construction lines for CAD is a nice concept as well. AutoCAD is a good example of what not to aspire to.

Note: I worked for HP for a summer helping out int their drafting department, but I haven't done any CAD since then (about 7 years ago), so my knowledge is a bit dated. AutoCAD might be a tolerable product now.

-jim

Re:CAD

[QuantumG]

Ummm.. how do you figure that mechanical engineers know how to code? I mean, yeah, maybe they know how to whip up a perl script or they can write some embedded C, but CAD software is not small - it's a complicated and ambitious project to make even simple CAD software.

Re:CAD

[Noksagt]

Ummm.. how do you figure that mechanical engineers know how to code?

Because some in academia do code. They have made intricate custon multi-scale finite element models and the like. There is F/OSS molecular dynamics, monte carlo, rendering, and other software. Until recently, CAD has been a very noticeable gap in the science/engineering programs in the F/OSS "library."

Re:CAD

[Anonymous Coward]

That's very naive of you...

A Simple, End-User Oriented Database

[Dragonmaster Lou]

Something along the lines of Access (though with a better DB engine that Access), FileMaker, Paradox, etc., would be nice.

It should use local files (so you don't need to have a server running, although that could be an option) and have an easy to use form layout system. I don't want to have to administer a database daemon, and I don't want to have to have to hand-hack code for a simple database.

I have mucked around for a while looking for something like this. The closest I've seen in Rekall, but it looks like it still needs to hook up to an external database of some kind, as best as I can tell.

Re:A Simple, End-User Oriented Database

[DrSkwid]

man join

man sort

man grep

that is all

Re:A Simple, End-User Oriented Database

[Isomer]

Like Glom or GNU enterprise. Both prefer to use postgres, but failing that, at least gnuenterprise can use sqllite for local database use (dunno about glom).
Both projects seem pretty good, they just need mindshare :)

Re:A Simple, End-User Oriented Database

[megabyte405]

GnomeDB Mergeant ?

Re:A Simple, End-User Oriented Database

[iantri]

Rekall can use XBase type databases with the XBSQL translator thingy. They are flat file databases.

It sounds more complicated than it is. But the answer is no, you don't need a server running.

Re:A Simple, End-User Oriented Database

[Gamasta]

I asked a friend this question about Access couple of days ago. He suggested I should get Apache HTTP Server, PHP and MySQL (in this order). I got them all installed and running in a W98 machine. I only do thinks locally, so I'm not concerned with security (root password is asdf if you want to know, and it's just to prevent me from doing something stupid. I access as another user with very limited privileges). But you have to learn quite a lot of stuff to get started. Three different programs. BTW, this should help: phpMySQLAdmin (in sourceforge). But I look forward to something such as open source access. It's the last thing that needs replacement in my box.

Cross-Platform Game Design Studio

[xanderwilson]

I know I'm probably dreaming, but I'd love something like an Open Source alternative to Director or Flash.

Alex.

Re:Cross-Platform Game Design Studio

[empaler]

Fucking... awesome... idea.

Re:Cross-Platform Game Design Studio

[Geoffreyerffoeg]

If OOo Impress gets a bit more polished and a bit more capable (even more than PPT) with respect to custom animation, especially motion paths and the like, it will probably suffice. It has an "Export to SWF" option. I've done some cheesy yet impressive Flash-like animations in Microsoft PowerPoint, and I know OOo can support everything I used.

Maybe Impress should be further developed with this goal in mind, in addition to presentation-making.

Re:Cross-Platform Game Design Studio

[josath]

He said "Game Design"....not "SWF Animation Design".

Re:Cross-Platform Game Design Studio

[NateKid]

This is very much needed. A few years ago, I actually switched from linux to windows in order to do flash work and have stuck with xp since then (could/should have gone the os x route, but I already had my boxen picked out and was doing .net work anyway). If linux had flash 2004 I'd definitely switch back, but until then I'm a sellout...

Re:Cross-Platform Game Design Studio

[Geoffreyerffoeg]

Sorry, didn't see the title. He did ask about Director and Flash, though.

Re:Automated

[Anonymous Coward]

See nget, pan, mldonkey, and pornview

WYSIWYG web design

[Johnny Mnemonic]

I've tried netscape etc, but I would sure like to see a web designer that will allow me to drag a picture anywhere inside a box, and build a table that positions it correctly relative to the other elements in the page. For instance something like this. If OSS can design a PhotoShop killer (GIMP), why not a truly WYSIWYG web designer?

Oh, and while we're dreaming, how about a desktop HIG standard? So each time I load a new distro I don't feel like I'm learning a new OS? While that's cool for hobbyists, it hurts corp adoption because Linux continually feels "unpolished." Why can't some consortium develop, decide on the lowest common denominator, and make it a standard that shrinkwrap developers and trainers target? And then you can leave the other stuff for preference panels.

Re:WYSIWYG web design

[name773]

paragraph 1: learn to use html via text editor, it gives you better control

paragraph 2: kde, gnome, etc. while different, they are all pretty consistent when taken alone.

Re:WYSIWYG web design

[vigilology]

"I would sure like to see a web designer that will allow me to drag a picture anywhere inside a box, and build a table that positions it correctly relative to the other elements in the page"

The '90s called. They want their table-based layouts back.

Re:WYSIWYG web design

[Anonymous Coward]

If OSS can design a PhotoShop killer (GIMP), why not a truly WYSIWYG web designer?

Open-source software is less likely to produce such a thing because people have more pride in their work when they are working on their hobby.

There's no such thing as a "truly" WYSIWYG web designer. Well, there is, but what you see is not necessarily what everyone else sees. The web is an inherently fluid medium, and it's impossible to have a tool that can create a layout that works under a variety of conditions. The best you can do is create a leaky abstraction that does the job in common situations and is liable to break often.

People who have pride in their work, and are qualified to write a "WYSIWYG" web designer, often don't because they know it's a fundamentally broken idea, and there really isn't a way of creating a decent layout without knowing the code and having experience in the field.

It sounds like elitism, but it really isn't. They just don't want to be responsible for crappy software that creates crappy websites.

Fortunately for you, a couple of companies have been paying developers to create such tools. Somebody has already mentioned Nvu, but Quanta also has a visual page layout mode.

Re:WYSIWYG web design

[Anonymous Coward]

you f*cking wanker. "learn how to do X" posts are indicative of someone with way too much time on their hands to learn "X".

Re:WYSIWYG web design

[kiddygrinder]

That is crap, a text editor CANNOT give you the same control as dreamweaver when building a table layout. I work 95% in notepad++ but whenever i'm doing layout, I still prefer to fire up the slow-as-shit behemoth to get it sorted.

Re:WYSIWYG web design

[name773]

so you want to do something quickly without learning how to do it, and you want the tool for free. have fun :P
also, considering that the people who could build the tool are most likely proficient in both html and c/c++/etc. they probably consider html to be a piece of cake and therefore would not have much motivation to build the program.

Re:WYSIWYG web design

[kiddygrinder]

I have yet to hear of any real reason to move from tables to css layouts, aside from the automatic entry into the elitist wanker class of web designers.

Re:WYSIWYG web design

[thulsey]

CANNOT give you the same control is a strong phrase to use about a tool that is basically giving you a front end to... um.... create markup. Which is text. That you can control. In a text editor. Or in wysiwyg html editor of choice.

That being said, I've always found that even if I use a wysiwyg html "layout tool" I've had to go in and manually tweak the markup by hand. In the end I just figured out what i wanted and let my fingers do the walking...

But, with table layouts going the way of the dodo, why bother?

Re:WYSIWYG web design

[thulsey]

Try this.

It ain't perfect, but it's free, multi-platform, and made by the guys that invented the standard.

It takes a little getting used to, as there are some powerful text-level commands you can use, as well, but the basic layout and viewing of the DOM are nifty.

Re:WYSIWYG web design

[kiddygrinder]

I suppose my language was a little strong there, however still feel it's correct; Which is easier, dropping '15px' in code, refreshing a browser to see how it looks, or dragging the side of a table (or div if you will) out a bit to get the right sizing. I've heavily invested time in doing it both ways and doing it in code just makes it harder.

Re:WYSIWYG web design

[PylonHead]

• faster page loads
• style decisions centralized in one file
• complete nightmare getting it to work across all browsers
• multi-column layouts rediculouly tedious to get working

Oh, f*** it. I would just stick with tables if I were you....

Re:WYSIWYG web design

[Anonymous Coward]

You probably haven't heard any reasons because going around calling people "elitist wankers" simply for using technology you don't understand isn't likely to get you lots of well-meaning advice. People are likely to just ignore you rather than try and explain things.

The main reason is separation of content and presentation. This has a number of benefits - for example, you can pay editors less if they can just care about the content wthout being experts in which browser acts funny given certain layouts etc. When styling is in an external stylesheet, it can be cached efficiently, leading to lower server load, lower bandwidth consumption, faster websites, etc.

Re:WYSIWYG web design

[Anonymous Coward]

Which is easier, dropping '15px' in code, refreshing a browser to see how it looks, or dragging the side of a table (or div if you will) out a bit to get the right sizing.

If you want exactly 15px? Typing it.

Re:WYSIWYG web design

[angst_ridden_hipster]

Reasons to use tables:

• CSS *sucks* for cross-browser compatibility. It takes a lot of work, and still will not support legacy browsers.
• Easy, since all the programs do it, and we've all been doing it since 1995. How tricky could <TR><TD> etc get?
• Multi-column layouts without intense pain. CSS can do this, but you have to play silly games.

Reasons to use CSS:

• Separates the design into a cachable file. Faster load times.
• Your HTML pages will be tiny. And probably usable on wierd devices. Again, faster load times.
• Skinnable sites, based on arbitrary criteria (server side or client side).
• Control. If you want pixel-level control, you can have it. At a price. And if you don't care about legacy browsers.
• Easier to change layouts generated by code. Change the stylesheet, not the code. I found this important when working on a site that statically built large numbers of pages and took a long time to create.
• Elitist Wankerdom. You know you want to be an Elitist Wanker like the rest of us. If you can't sling acronyms like CSS and XSLT (oooh!), you just have to face up to the fact that you suck.

There you go folks. A handy guide. Now back to your regularly scheduled quibbling.

Re:WYSIWYG web design

[kiddygrinder]

Probably shouldn't bother rebutting, since you kinda agreed with me but: I have never made a page that was more than 5k bigger in tables than the other guy I work with can do it in css(discluding stylesheets, since they cache) Whether I'm in php or asp, the entire layout is in header.asp and footer.asp which are included into every file, with a bunch of text with the least amount of markup possible in the middle. This makes it pretty trivial to do site-wide layout changes.

Re:WYSIWYG web design

[Johnny Mnemonic]

paragraph 1: learn to use html via text editor, it gives you better control

Thanks, I'm aware of that. I could also learn to play the violin so I don't have to listen to the radio. Oh, that's right--it's not something that I want to spend an hour a week on for the next year.

I don't create lots of websites. If I did, I'd learn HTML. Rather, I would like to throw up the odd one or two page "here's my pics" sites, but I don't want to have to fuss with tables every time. I just want to spend an hour at it.

Re:WYSIWYG web design

[Anonymous Coward]

CSS *sucks* for cross-browser compatibility. It takes a lot of work, and still will not support legacy browsers.

I think most people who make a big deal about this forget how much time they've already spent working with table layouts. Remembering that certain optional closing elements will screw up certain browsers if you miss them out. Remembering how to keep track of the colspans and rowspans. Remembering how to track down the gaps that somehow appear in some browsers when you didn't think you changed anything of importance. Remembering the time when one half of a submit button was clickable and one half wasn't, even though your table was nested perfectly. All kinds of weird stuff that most web developers do without even thinking just because they've done it for so long.

I'll admit that there are certain really odd things that crop up with CSS too, especially in Internet Explorer, but really, it isn't much worse than when everyone was first starting out with table layouts.

Multi-column layouts without intense pain. CSS can do this, but you have to play silly games.

What's so hard about setting a width and floating it off to the side? For the vast majority of layouts, CSS is straightforward. It's harder when you get into the more complex layouts, but the same applies to table layouts as well. Oh, and if Internet Explorer supported the six year old CSS 2 specification, it would be even easier, as you could just use display: table-cell.

Control. If you want pixel-level control, you can have it.

CSS does not give you this. HTML certainly doesn't give you this either, the whole idea of "control" is antithetical to the web. In fact, a core part of CSS is designed around the lack of control - you know the 'C' in 'CSS' stands for "cascading"? That's the mechanism by which different stylesheets, including user stylesheets, combine to provide the final appearance. If you want control, the web isn't the right medium for you.

Re:WYSIWYG web design

[Anonymous Coward]

I have never made a page that was more than 5k bigger in tables than the other guy I work with can do it in css(discluding stylesheets, since they cache)

With all due respect, either the guy you work with isn't very good with HTML & CSS, or you don't create very complex pages. In my experience, moving the presentation crud into stylesheets reduces HTML documents by a tremenduous amount. The various volunteer redesigns for popular websites (e.g. Microsoft, Slashdot) show a similar reduction.

Whether I'm in php or asp, the entire layout is in header.asp and footer.asp which are included into every file

You do realise that the header and footer includes constitute part of the HTML that is sent to the browser don't you? In any case, if the complicated stuff can go into the header and footer, it really sounds like you are just creating simple pages and nothing at all complex.

Re:WYSIWYG web design

[I_Love_Pocky!]

Oh, that's right--it's not something that I want to spend an hour a week on for the next year.

I have a little bit of inside information for you: HTML is not the kind of thing that would take an hour a week for an entire year to learn. Unless you aren't very swift (my appologies if you are), I would think you could learn enough HTML to do what you seem to want to do in between 1-2 hours.

I don't create lots of websites. If I did, I'd learn HTML. Rather, I would like to throw up the odd one or two page "here's my pics" sites, but I don't want to have to fuss with tables every time. I just want to spend an hour at it.

Sounds like you could get by just fine with the webpage editor that comes with Mozilla for something like that.

Having said all that, I completely agree that something that would rival Dreamweaver would be great to see in OSS. It isn't a tool for amatures to avoid learning HTML. It is a tool designed for professionals to save time. And Dreamweaver does a great job at it.

Re:WYSIWYG web design

[Anonymous Coward]

For most layouts you can pretty much get your design dead on, before even seeing what it looks like in the browser. It's not hard really, you just need to understand what is going on, which is mostly experience. I won't get into which is better, but I know which I prefer, and I can use any old text editor to do it with!

Re:WYSIWYG web design

[thulsey]

Excellent point.

Using a GUI to "draw" a layout is much more intuitive, and easier to do (providing that your motor skills are up to the challenge...). There's a reason we don't write postscript files by hand and use WYSIWYG vector drawing tools instead -- it's easier, and simplifies a potentially complex process.

Of course, if you are getting your content generated dynamically, things change.

Building your table in a while loop, for example. Then all your control happens in the markup. Not as easy, but you have just as much control.

I'm not really arguing with you. I agree, it's way easier and convenient...er to use a tool like Dreamweaver or Nvu to slam out content. It's just not always possible or practical.

Now what you need is a "live preview" editable wysiwig browser/editor that could grab content from a db or external link, preview and render with CSS and still be "editable." I tried Amaya thinking it might work that way, and couldn't even get it to render positioned divs correctly. Dreamweaver has "Live preview" but you can't really edit the generated stuff that well...

Re:WYSIWYG web design

[Anonymous Coward]

Using a GUI to "draw" a layout is much more intuitive, and easier to do

That is true if and only if we are talking about static layouts. The web isn't static. Is it intuitive and easy to "draw" a layout when it could change depending on font size, resolution, canvas size, etc? How is any sort of visual "drawing" supposed to capture those kinds of subtleties? It can't. That's why so-called "WYSIWYG" editors are simply not up to the job of designing a web page.

Re:WYSIWYG web design

[trewornan]

when building a table layout

I don't blame people for still using the Killer Webpages (aka "table layout") paradigm - I do it myself sometimes, mainly because that's the way I thought for so long it's easier as a fall back to when CSS gets tricky (I know I shouldn't but the hell with it).

But, to propose this as the basic layout strategy for a new WYSIWYG web design application . . . I'm stunned that there are still people with such a backwards outlook.

Re:WYSIWYG web design

[jonadab]

> paragraph 1: learn to use html via text editor, it gives you better control
> Thanks, I'm aware of that. I could also learn to play the violin so I don't
> have to listen to the radio. Oh, that's right--it's not something that I
> want to spend an hour a week on for the next year.

Huh? If you spent an hour a week learning HTML, you'd know HTML very well
in a month. It takes about thirty minutes to learn enough HTML that you never
want to use a WYSIWYG HTML editor again. At that point you'll still keep a
reference handy (like the one at w3schools.com or the one at htmlhelp.com),
but it's still faster and easier than fighting with a WYSIWYG editor.

Oh, and don't bother with HTML 4.0 at this point; just skip straight to XHTML.
It's actually simpler and easier, if you don't have preconceptions in your
head from a prior knowledge of older versions of HTML.

Re:WYSIWYG web design

[jonadab]

> Having said all that, I completely agree that something that would rival
> Dreamweaver would be great to see in OSS. It isn't a tool for amatures to
> avoid learning HTML. It is a tool designed for professionals to save time.
> And Dreamweaver does a great job at it.

Except, it doesn't save time: it *costs* time, because as soon as you want
to make a nontrivial change, you have to redo the whole thing from scratch.

Re:WYSIWYG web design

[jonadab]

> Which is easier, dropping '15px' in code, refreshing a browser to see how
> it looks, or dragging the side of a table (or div if you will) out a bit
> to get the right sizing.

Your problem is, you're writing the code the same way the WYSIWYG editor would,
which is to say the wrong way. 15px is wrong for a very significant percentage
of the users viewing your page, because (gasp) they aren't all viewing it on
your computer (with your resolution and window size and other settings).

Specifying things in pixels is just one of numerous glaring mistakes WYSIWYG
editors invariably make that cause their output to need to be rewritten from
scratch before it can be usefully deployed on the web. If you don't want to
bother rewriting it, you end up telling your users, "Please view this page in
this specific version of this specific browser on this specific version of
this specific operating system with these specific color settings in your
theme, with your font size set to this specific value and your resolution set
to this specific size and your taskbar in this position and this size, and
maximize your window, and make sure you have these fonts installed, and set
your preferences to these specific values..." and then if the user sets the
GUI to "Classic" instead of Luna, it throws everything off by three pixels
and your whole layout is ruined.

Of course, no two of these websites have exactly the same requirements. The
first couple of times users see these kinds of requirements, some of them
attempt to comply (if they know how to change such things, which many don't),
but after a dozen or so sites ask them to *yet again* tweak their settings,
they get tired of it and just go find another site.

Yet, somehow, there are websites out there that manage to scale, so that they
look okay at 640x480 and yet still look good at 1600x1200.

Re:WYSIWYG web design

[jonadab]

> Control. If you want pixel-level control, you can have it. At a price.
> And if you don't care about legacy browsers.

My usual standards for legacy browsers (basically, anything over two years old)
is that the page be legible and usable. If the layout doesn't look quite as
intended, I don't fret over it, as long as everything's there, doesn't overlap,
and so on.

These are my same criteria for less than 24-bit color: it has to be legible
(down to 16-color mode, and also 8-bit greyscale; I do not attempt to support
1-bit monochrome); it doesn't have to look great at the low end, though. And
with image-loading turned off, again, I use the same criteria: it has to be
legible and usable, but it doesn't have to look like it was intended to be
viewed that way. And with scripts disabled -- it has to be possible to
navigate the site, but if the user has to type things in by hand that the
scripts would have automated, so be it. And so on.

This allows a site to be _enhanced_ by newer technology without _requiring_ it.

Re:WYSIWYG web design

[Anonymous Coward]

dropping '15px' in code

People like you are the reason that people like me are losing their eyesight.
I run my 19-inch Monitor at 1600x1200, and this small-pixel-sized text makes things just so hard to read.
There is no excuse for using pixels any more; use points (pt), millimeters (mm), or inches (in), but not pixels.

Re:WYSIWYG web design

[Anonymous Coward]

try and explain things

"try to explain".

Re:WYSIWYG web design

[I_Love_Pocky!]

because as soon as you want to make a nontrivial change, you have to redo the whole thing from scratch.

Maybe you don't know what you are doing. I have a clear idea of how what I do in Dreamweaver will affect the source. Even if it does do something I dont like it takes very little work in the source to remedy it. It certainly doesn't mean that I have to redo the whole thing from scratch.

If it costs you time, don't use it. For the rest of us who know what we are doing, we are going to keep using it. Remeber one thing: HTML is super easy, and doing it by hand does not make you hardcore. I can "code" HTML in my sleep.

Re:WYSIWYG web design

[angst_ridden_hipster]

I'll admit that there are certain really odd things that crop up with CSS too, especially in Internet Explorer, but really, it isn't much worse...

I guess it all comes down to requirements. When you have clients who demand things work in IE 5.0, you really worry about bad CSS implementations. Most table issues have been worked out by browsers with versions > 4, or only happen with deep nesting.

Generally, I work with layouts that a single table would handle, or, perhaps, one level of nesting. In most cases, it's trivial to do in CSS. In some cases, it's quite difficult (see below).

What's so hard about setting a width and floating it off to the side? For the vast majority of layouts, CSS is straightforward.

Once again, what's straightforward is relative. For example, making a three-column layout with a header div and a footer div, and which centers in the user's browser window, requires a lot of silly things like negative padding. Sure, you can say "but don't do that!" Unfortunately, my job is usually to implement someone else's design, and that someone else works with the person who pays the bills.

CSS does not give you this. HTML certainly doesn't give you this either, the whole idea of "control" is antithetical to the web.

Of course. The whole idea behind HTML was easy, device independent markup. Now welcome to the real world of clients. You can build them what they want, or you can watch them go to some other developer with their money.

The fact is, CSS gives you pixel-level control within some fairly rigid constraints. If you're working with IE 6 and Mozilla, you can provide a precise layout (assuming, as you point out, that the user does not override it).

you know the 'C' in 'CSS' stands for "cascading"

Golly. And here I was, stupidly thinking it stood for "Condescending."

If you want control, the web isn't the right medium for you.

Nice sentiment, and, on a technical level I agree. If I could, I'd be spending all my time doing the interesting work in the back-end using well-designed frameworks, designing a efficient database schemas, and then just throwing simple tabular interfaces on 'em. But once again, welcome to my real world. Clients hire me to put their applications online. They have strong ideas about how things should look, and it's not my place to tell them that they can't do what they want. I can give suggestions, I can make recommendations, but in the end, the design is their decision. If it's a truly technical issue (i.e., "but people will mis-use that application to send spam!), I can insist. When it comes to layout, I can't.

CSS gives me pixel-level control for an agreed-upon requirement set. That really sums it up for me. CSS is a tool that saves me time and makes for faster loading pages. It doesn't solve all problems for all situations. And frankly, I can be nondogmatic. I use a mix of both CSS and tables -- whatever works best and requires the least effort.

Re:WYSIWYG web design

[PylonHead]

There are definitely plusses and minuses. On the plus side,

1) You can really get in there and modify the look of lists and paragraphs.

2) We do mostly content management solutions where the clients are going to be editing the pages, so it's helpful to setup all the styles outside, and have them apply to what the editors create.

This definitely includes navigation, which, if you do with styles, can change dynamically. If you do it with graphical rollovers, you either have to resort to auto generated graphics, or stick with the nav you have.

3) Your pages become cleaner and thus easier to edit.

4) We also use templates for all our boilerplate code, but sometimes we have multiple back ends (some basic pages in php, combined with an ecommerce package). A single style sheet can provide a uniform look across the systems.

On the bad side:

1) You have to learn something new

2) We still have to just through some stupid hoops to make it work across IE 5.0 and other non-compliant browsers. It's constantly getting better though as older browsers die off, and the new ones get with the plan.

3) CSS need to be able to handle multi column layouts. I know there are entire websites dedicated to this stuff. There just don't seem to be any good answers. (At least given the 3 or 4 days I wasted trying to get this to work).

Overall, I'm really anal about abstracting everything I can... seperating content from presentation. So for me it's worth it, flawed as it is.

Re:WYSIWYG web design

[mdavids]

CSS *sucks* for cross-browser compatibility. It takes a lot of work, and still will not support legacy browsers.

CSS-based pages render perfectly legibly in pre-CSS browsers. Not pretty, but legible. I think you're talking about those few browsers from Microsoft and Netscape that claim to support CSS, but in reality deliberately misinterpret CSS with the intention of breaking the standard, and forcing authors to create "This site best viewed with..." websites. I don't see how you can blame the standard for the consequenses of a couple of rogue companies trying to sabotage the web, or expect the standard to adapt to support this "legacy" behaviour.

Easy, since all the programs do it, and we've all been doing it since 1995. How tricky could <TR><TD> etc get?

You've obviously never had to maintain somebody else's tables-based code, or you have a Rain-Man-like capacity for knowing precisely where you are when you're a dozen nested tables deep into a page.

Multi-column layouts without intense pain. CSS can do this, but you have to play silly games.

What's so hard about:

div#column1
{
width : 70%;
float : left;
}
div#column2
{
width : 30%
float : right;
}

Re:WYSIWYG web design

[Anonymous Coward]

When you have clients who demand things work in IE 5.0, you really worry about bad CSS implementations. Most table issues have been worked out by browsers with versions > 4, or only happen with deep nesting.

I still think you are wearing rose-tinted glasses when looking at the table layouts. A lot of the problems with CSS only crop up in complex layouts too. You still have to remember various browser oddities when dealing with table layouts, and since tables weren't designed for page layout, things aren't guaranteed to work forever. There's already been backwards-incompatible changes for table rendering that have tripped up the people relying on certain behaviour.

For example, making a three-column layout with a header div and a footer div, and which centers in the user's browser window, requires a lot of silly things like negative padding.

Huh? No it doesn't. Three divs, float them to the left and set their widths. Put a header div above, put a footer div below and clear it. Wrap the whole thing up in a container div that centres it and you are done. And it would be even simpler if Internet Explorer implemented the six year old CSS 2 specification that Microsoft helped write.

Of course. The whole idea behind HTML was easy, device independent markup. Now welcome to the real world of clients. You can build them what they want, or you can watch them go to some other developer with their money.

You are implying that "what they want" is control, so the first option is not an option at all. As a web developer, it's not in your power to give them control. You can mislead them in an attempt to keep them from going to more ethical companies, but you can't give them control.

The fact is, CSS gives you pixel-level control within some fairly rigid constraints. If you're working with IE 6 and Mozilla, you can provide a precise layout (assuming, as you point out, that the user does not override it).

You are talking as if "override" is some big button that a user presses specifically to screw with you. I'm talking about their normal settings not being condusive to particular designs.

But once again, welcome to my real world. Clients hire me to put their applications online. They have strong ideas about how things should look, and it's not my place to tell them that they can't do what they want.

Yes, yes it is. If a client asks you for something that is impossible, it's not ethical to pretend that it is possible and hope you get away with it. A responsible person would explain to them the issues involved. Virtually every time, they'll come back with "keep it as close as you can to what we want without breaking things for people".

Re:WYSIWYG web design

[angst_ridden_hipster]

Wow. So I've gone from giving my opinions on the relative strengths and weaknesses of CSS and tables, to being unethical, irresponsible, and promising the impossible. Gotta love Slashdot.

In any case, I think you're reading a lot into my comments that is not what I intended to express. Fine, I'm not the most articulate writer in the world. Whatever. Fundamentally, we're actually pretty close to agreement, and if you weren't so anxious to paint me as some kind of an ignorant and deceptive idiot, I'd be interested in continuing this conversation to clarify those points. But, as it happens, you is and so I ain't.

Re:WYSIWYG web design

[angst_ridden_hipster]

I don't see how you can blame the standard...

It wasn't really my intent to blame the standard. I just have the experience of having to implement pages that work on those legacy browsers. So I have to adopt a hybrid approach.

Unfortunately, those "couple of rogue companies" you mention happen to completely dominate the browser market. With any luck, Moz/Firefox/Konq et al will change that. But for now, like it or not, I have to be able to work with IE.

You've obviously never had to maintain somebody else's tables-based code...

Point taken. People can make some pretty ugly tables. I've seen 'em and loathed 'em. And, getting back to the grandparent's point, some programs can generate some vile HTML.

What's so hard about...

Consider a case where you have three-column interface, different background shades for each column, all with a header and footer div. Oh, and it all wants to be centered in the browser window. Sure, it should be as easy as the example you provide, but IE has serious problems with floats within, say, the center column (i.e., subcolumns). It's not a problem with the standard. It's a problem with the implementations of the standard. But I'm stuck supporting that implementation.

Re:WYSIWYG web design

[jonadab]

> Maybe you don't know what you are doing. I have a clear idea of how
> what I do in Dreamweaver will affect the source.

I know *exactly* how what I do in Emacs will affect the source :-)

> Even if it does do something I dont like it takes very little work in
> the source to remedy it. It certainly doesn't mean that I have to redo
> the whole thing from scratch.

Perhaps WYSIWYG editors have improved since I last looked at them, but
every time I see a web page created in one... I see evidence to the
contrary. Maybe you're one of a few rare people who can actually make
them work properly? Dunno, but vanishingly close to 100% of the web
pages created with them are such utter crap, they need to be rewritten
from scratch just to get them to scale to a resolution other than the
one the designer was using. If you can manage better, good for you.

> If it costs you time, don't use it.

I don't.

> For the rest of us who know what we are doing, we are going to keep using it.

I don't think it's reasonable to conclude that everyone who uses it knows
what they're doing. I'm certain that's not so (as with most software).

> Remeber one thing: HTML is super easy, and doing it by hand does not make
> you hardcore. I can "code" HTML in my sleep.

That was rather my point: HTML is so easy to do by hand, wanting a GUI
tool for creating it is... just plain bizarre. A text editor with a
handful of macros is much more efficient. Admittedly, what I use is not
so much a handful of macros as almost a major mode that I've put together
out of bits and pieces over the years, and that integrates with and
piggybacks on cperl-mode, but you wouldn't have to have all of that to
be more efficient than a WYSIWYG editor; half a dozen macros would do in
a pinch, I would think.

There are even people who claim Notepad is a good HTML editor, but you
will note that I don't go that far. Editing HTML in Notepad would be
quite tedious, because of the lack of macro support; you'd have to type
every character of both the opening and closing tags for each element,
which would be rather more typing than is strictly necessary. If that's
what you're comparing to, then I can see why you prefer FrontPage.

As far as being hardcore, I didn't intend to imply that, sorry for any
confusion; indeed, computers aren't really even my strong subject (though
they're not my weak subject either). But as you point out, HTML is a very
long way from hardcore.

Re:WYSIWYG web design

[I_Love_Pocky!]

Dunno, but vanishingly close to 100% of the web pages created with them are such utter crap, they need to be rewritten from scratch just to get them to scale to a resolution other than the one the designer was using.

How would you know that? I'm sure you only notice the ones that are done poorly. Did it ever occur to you that some of the websites you don't think are crap were constructed using a GUI tool?

I don't think it's reasonable to conclude that everyone who uses it knows what they're doing.

I don't either.

A text editor with a handful of macros is much more efficient.

I just disagree on this. For a standard web page, I think a GUI tool for laying out tables is far quicker than using a text editor (particularly if you have to make changes to an existing table). To be fair I haven't done web development professionally in five years. The only stuff I do now is for my own personal websites. For my own sites I do all of my web work in vim, because GUI tools don't lend themselves well to php development. So, even though I see the value in a tool like Dreamweaver, I don't actually use it anymore (But I highly recommend it to anyone who already knows HTML).

I can see why you prefer FrontPage.

FrontPage is completely worthless.

I'd just like to apologize if you have taken any of my statements personally. I can certainly see where you are coming from, but I am just trying to dispel the idea that the only people who use GUI HTML tools are people who don't know/are too lazy to learn HTML. This simply isn't the case.

Re:WYSIWYG web design

[kiddygrinder]

You're wrong, even in a fluid design you still have fixed width elements, duh. Do you even know what you're talking about?

1001 vertical market apps

[leonbrooks]

A development kit that helped people migrate from Pronto, DataFlex and the like to something open would be good. Likewise a more comprehensive VB-to-real-language translator that did stuff like retrieved the dialogs and forms would be most helpful.

QuarkXPress/InDisign replacement

[rekrutacja]

DTP on linux is simply impossible. Scribus doesn't work (yet?) for most of us...

Done and done.

[sethadam1]

Nvu is your answer.

Re:Done and done.

[JabberWokky]

...and the new version of Quanta Plus.

Good Project Management software

[Tye_Informer]

Until my manager can create/update project plans under Linux, it will not even be considered as a replacement on the desktop.

Re:Good Project Management software

[Darth_Burrito]

I looked at dotproject for a while. Never used it in a real world setting, but others have.

Re:Good Project Management software

[insensitive claude]

There are several in development. As usual, the closed file formats of the defacto standard are going to be an even bigger hurdle than usability in growing these apps.

Collaborative calendar app...

[cjpez]

... that's cross-platform and lets you reserve conference rooms, schedule meetings, etc, etc, etc. Haven't found anything quite like that yet, not that can be used on coworkers' windows machines too, anyway.

Re:Collaborative calendar app...

[acaird]

I agree and am suprised that no one else mentioned this. We have decent (not great, but passable) word processors and spreadsheet programs. Good web browsers. Very good networking and server level stuff. But when people talk about why they use Windows, it's because of Outlook and the calendar. If there was a "firefox" of calendaring it would be really nice. It's more than a client though (unless you can manage to get Exchange calendaring working), so you have to get the server, and it has to operate with the huge MS Outlook base out there.
So, um, good luck.
.

Re:Collaborative calendar app...

[Noksagt]

But when people talk about why they use Windows, it's because of Outlook and the calendar.

Have you seen Evolution? I'd say it was "there."

If there was a "firefox" of calendaring it would be really nice.

Eventually, there will be mozilla sunbird. Still has a long way to go, though.

Re:Collaborative calendar app...

[MrZaius]

Is there any reason that sort of application should be run on the client machine? It seems like any decent web-based calander could have a username set up for each conference room. What else would it take, and is there any reason you couldn't slip a college student a C-note or two to throw the other stuff into a preexisting a web-based calander?

That would also mean you'd only have one node to fix, when new security issues come up. Don't want to upgrade Outlook on all 50-50,000 nodes every year or three? Put it all on one or two web servers.

Re:Collaborative calendar app...

[cjpez]

There's a lot of things that HTML isn't good for, and while it's certainly possible to write a good HTML-based calendar app, I feel there are very good reasons to have a dedicated client. It's very nice in a calendar app to be able to click on 2:00 and drag until 3:30. It's nice to be able to right-click on things and have appropriate context menus to pop up. It's wonderful to be able to not have to worry about the fundamental sessionlessness of HTML, and having to deal with HTML form input processing and state maintenance (not that you can't just have a Session ID and store all your data there, but still, it's a pain). It's nice to have an app which can pop up a reminder 5 minutes before your meeting is about to come up, without having to resort to hokey meta refresh tags. HTML formatting isn't very well suited for doing things like having multiple blocks of time laid out on a calendar with what you'd expect to be clickable being actually clickable (not that it's impossible, especially with DHTML, etc, but still).

I could probably go on for awhile. Again, I'm sure that you could write a good calendar app on a webpage. But I don't think it would be nearly as good as an actual client. When you write a client you've got the power to construct protocols that make *sense* for the application you're designing. You've got a great deal more control in general. I know some people are of the opinion that everything should be a webapp, but I disagree. I find webapps for this kind of thing to be, in general, quite clunky and nonintuitive.

You could argue, of course, that many of these concerns could be addressed by using Java applets or something similar on the page, but once you start doing that you may as well just take the extra step and write an actual Java based client or the like...

Re:Collaborative calendar app...

[cjpez]

Have you seen Evolution? I'd say it was "there."

The last time I used Evolution, the calendar was only designed for a single user... Have they added in the kind of collaborative system that I'm more concerned about? (scheduling appointments, reserving rooms, selecting a group of people and saying "when are all these people free?", etc?) Just curious, because things may have changed since last I looked.

Re:Collaborative calendar app...

[cjpez]

At a company I used to work for, we used the Netscape Calendar system (this was back in the 4.x days), and it was actually a really really good system. It had Windows and Linux clients too, which was great because the company was pretty much split 50/50 that way. I've never really run Outlook's calendar myself, as I only run Linux, but I really do miss having something like that on my Linux box. The Netscape product seems to have basically disappeared (I think it was on the way out when I joined up with the company, even).

The company I'm at now just started a big transition over to Outlook and all that, and I know the calendaring system was one of the factors affecting the decision. All the internal mail stuff used to be handled by OSS boxes, now it's Outlook. (Granted, the company is like 95% windows internally at least, so it kind of makes sense for them to be doing it, but it's painful for those of us in the 5% who don't even really live on the internal corporate network.)

Re:Collaborative calendar app...

[Noksagt]

Yes! There is a Ximian connector for MS Exchange. Collaborative scheduling is possible.

Re:Collaborative calendar app...

[cjpez]

Isn't Ximian Connector non-free, and doesn't it require an Exchange server on the backend?

Re:Collaborative calendar app...

[Gamasta]

cross-platform -- perhaps php? Take a look at this webcalendar

http://sourceforge.net/projects/webcalendar/

Usese php and MySQL. I find it quite nice, but I'm not sure it has everything you require. Then again, this is open source. You can just add the features or branch the project (but then people might say your project is dying, and Netcraft might confirm that).

Re:Collaborative calendar app...

[Noksagt]

Now Novell Connector, it is GPLed. It was made to work with Exchange servers (acaird said in the first post I replied to that Outlook (and hence Exchange) kept people on Windows & I did say it was for Exchange). However, Evolution need not connect to an Exchange server: it supports iCalendar and Groupware too.

Re:Collaborative calendar app...

[cjpez]

Hm, fun, I'll have to check it out then. Is iCalendar a Mac thing?

Re:Collaborative calendar app...

[cjpez]

Well, as I said elsewhere in the thread, while I'm not opposed to a web-based calendar system, I'd much prefer having a dedicated client. Apparently Ximian Connector's been GPLed and released as Novell Connector now, though, so perhaps what I'm looking for *does* exist.

Re:Collaborative calendar app...

[cjpez]

Oh yeah... That's the stuff.

It looks like it's still fairly tied into an Exchange background, which kind of sucks; I'd really love to have an OSS backend server as well. At least this'd be a way to run a client in Linux though.

Re:Collaborative calendar app...

[namekuseijin]

We have decent (not great, but passable) word processors and spreadsheet programs.

I don't think the likes of OpenOffice or Gnumeric are merely "passable". Gotta love Gnumeric.

Good web browsers.

Ah! Media theses days!... i was under the impression the Mozilla Firefox was the best browser we had today, bar none. Thanks for clarifying that up and put it to where it belongs.

I'm curious, though, as how low would you rank IE, then....

But when people talk about why they use Windows, it's because of Outlook and the calendar.

I call it BS. People use Windows because that's what came in the box, because it's what everyone else uses and because they're most likely not even aware that alternatives exist. Heck, i don't even think the regular Joe is aware of such words as "Microsoft", "Windows" or "Internet Explorer". He likely just refers to "internet" or "word".

But anyway, a calendaring solution is coming from the Mozilla Foundation and couple that with Firebird, and you have and Outlook killa.

Or you can use the more robust Ximian Evolution software. :)

Re:Collaborative calendar app...

[Noksagt]

Look at OpenExchange.

Re:Collaborative calendar app...

[Noksagt]

Mac is pushing iCal, but it is an open standard & you just need a webdav-enabled server on your backend to use it.

Re:Collaborative calendar app...

[Grishnakh]

The company I'm at now just started a big transition over to Outlook and all that, and I know the calendaring system was one of the factors affecting the decision. All the internal mail stuff used to be handled by OSS boxes, now it's Outlook. (Granted, the company is like 95% windows internally at least, so it kind of makes sense for them to be doing it, but it's painful for those of us in the 5% who don't even really live on the internal corporate network.)

Your company could use my company's simple solution: for everyone that normally uses a Linux desktop, just shell out a mere $2700 for a nice IBM Thinkpad for them to use to read their email and do calendaring.

Or, if they're smart, they could use WTS/Citrix, so a small number of Windoze servers could provide these services to all the Linux-using employees.

Re:Collaborative calendar app...

[cjpez]

Your company could use my company's simple solution: for everyone that normally uses a Linux desktop, just shell out a mere $2700 for a nice IBM Thinkpad for them to use to read their email and do calendaring.

Ugh, that's really really annoying though. The company that used to use Netscape Calendar ended up moving to Exchange shortly before I left, and they did have the solution of giving people a windows PC to do email/calendar/etc on (which was a hell of a lot cheaper than $3k laptops), but it's just lame. Or at least it was for me, because I have yet to find any email client that gives me any reason whatsoever to move away from mutt, etc. In general having things on two different boxes is just really lame anyway, though. I like being able to copy-n-paste things into my email, which is difficult if what I'm copying is running on a different box, etc. I shouldn't have to either transfer files or set up Samba shares just to attach a file to an email easily.

Regardless, email itself isn't a problem, because you can configure Exchange to do your ordinary average IMAP, etc, so any client that understands IMAP can talk to it. I think what they're doing for calendar type stuff is maybe giving the Ximian/Novell Connector thing a try.

Re:Collaborative calendar app...

[Grishnakh]

Ugh, that's really really annoying though. The company that used to use Netscape Calendar ended up moving to Exchange shortly before I left, and they did have the solution of giving people a windows PC to do email/calendar/etc on (which was a hell of a lot cheaper than $3k laptops), but it's just lame.

Hey, I didn't say it was smart or elegant, it's just what my company does. It's also pretty funny that they shell out nearly $3k for a laptop (when sub-$1k ones are available these days), but then have implemented cost-cutting measures in many other areas. Oh well, at least I get a laptop to carry around in case I need to.

Personally, because we also need to use MS Office files now and then, I like the WTS solution. I use it at work (I'm not sure why they still have WTS now that everyone has a laptop); I prefer to leave my laptop at home instead of lugging it around, and then I just do my Windoze crap in a window on my Linux box. And I never have to waste time installing hotfixes, rebooting, etc.: all the administration is done by the IT department.

Re:Collaborative calendar app...

[mbrinkm]

You haven't gone to Office2003 yet have you?

The reason I say that is Outlook 2003 will not allow you to share information without an Exchange server or someother connector and server software. I have looked for available options, if you want to share your calendar, contacts, etc in 2003 you need a server of some kind.

Re:Collaborative calendar app...

[cjpez]

Don't really know what you mean re: office 2003, etc... I'm on Linux so I haven't gone to an Office of any sort.

As to your other point, yeah, I'm down with the necessity of a server. I've always seen that as an integral part, and that's the bit which is most lacking in the OSS community I think. Some other people have pointed out Ximian/Novell Connector now, which looks like it's a good clientside app, so that's good on that front, but I'd like to see an OSS calendar server that's good as well.

Re:Collaborative calendar app...

[mbrinkm]

I thought you were talking about MS Outlook and sharing calendars, etc. My bad.

I've had a hell of a time explaining the need for a server to small-business owners when they've upgraded to Office 2003. Outlook used to be decent for collaboration on its own.

As for a calendar server system, have you tried TUTOS or eGroupware? Without knowing the details that you are looking for in a calendar server, these two could fit the bill.

Re:Collaborative calendar app...

[cjpez]

As for a calendar server system, have you tried TUTOS or eGroupware?

I'm rather adverse to using a web application for this kind of thing, for reasons I've enumerated elsewhere on the site. I've played with TUTOS in the past, but not eGroupware.

Re:Collaborative calendar app...

[cjpez]

Er, for reasons I've enumerated elsewhere IN THIS THREAD rather.

... and I even hit "preview" first.

_all_ FOSS software "still needs to be written"

[LeninZhiv]

The simple fact that the lion's share of a given niche is clearly held by one F/OSS offering (think LAMP), does not mean that there aren't parts of every F/OSS application that cannot be improved upon by anyone. Look at the bug trackers and todo lists of the projects that interest you; contributing, even to a well-entrenched project, is not impossible!

Dreams of geek celebrity status aside, making Linux/Apache/OO.org/YourFavouriteProject better does just as much for 'advancing the cause' as starting a new "killer app" from scratch does (and in 99% of cases, probably more).

How about...

[r3m0t]

CAD
Good IDEs
Movie/animation editing
Professional DVD menu editing (Look, it's complicated)
Graphical LOGO
Macromedia Flash or Shockwave editor (or editor of something similarly good)
Vector-based drawing

Re:How about...

[Noksagt]

Good IDEs

This is the most surprising request. What IDEs do you like? What is it about KDevelop, Anjuta, eclipse, and others that don't meet your needs?

Re:How about...

[Anonymous Coward]

Graphical LOGO

http://pylogo.org/

Re:How about...

[jonadab]

> Good IDEs

As opposed to Emacs, which has had more than ten times the functionality of
its nearest competitor for decades? Name a feature it lacks, that *any*
other IDE has.

Granted, it also has the "learning curve" feature in a significant way.
That could be greatly improved by a configuration that mass-rebinds all
the keys to match what users of other software expect... but this would
be a huge undertaking, because it would have to rebind the keys in every
important major and minor mode.

But, the _functionality_ is there.

Re:How about...

[evilviper]

There are a few "CAD" programs.

Tons of "IDEs" exist.

Several "Movie/animation editing" programs exist.

DVD-menu editing is quite a fringe item. A simple script could whip-up a nice generic menu for whatever set of videos you want to master.

I have no idea what you mean by "Graphical LOGO", unless you want a splash screen.

I've personally used a couple different open source vector-drawing programs. I've seen samples of incredibly complex images created with them.

Re:How about...

[r3m0t]

> DVD-menu editing is quite a fringe item. A simple script could whip-up a nice generic menu for whatever set of videos you want to master.

I was thinking of the advanced sort with animations, cool graphics, switches (subtitles on/off) and so on. The sort of menus you can see in most commercial DVDs. (Follow that link to doom9.net (unless that's .org, I forget) to see what I mean.)

I understand that professional graphics artists have Mac OS X as their OS of choice. They would be easy to convert... right?

Re:How about...

[r3m0t]

Well, can you graphically create Qt or Gtk+ dialogs?

If so, why the hell is it an Emacs addon/plugin/thingy?

Re:How about...

[evilviper]

I was thinking of the advanced sort

Exactly my point... It is more of a fringe item. Not something most people will need.

Re:How about...

[jonadab]

> Well, can you graphically create Qt or Gtk+ dialogs?

Dunno. I admit, I haven't done any GUI programming. But it wouldn't surprise
me if there were a module for that. (I happen to know that there's a WYSIWYG
module for editing TeX documents, another thing I've never used, and which
sounds implausible to people who don't grok the Emacs mindset, namely, "of
course there's a module for that".)

> If so, why the hell is it an Emacs addon/plugin/thingy?

A module. Because, every high-level feature in Emacs is provided by a module.
This is why autoloading is possible, which allows Emacs to consume quite small
amounts of RAM in practice, despite gargantuan amounts of functionality being
provided.

There may not be a module for graphically creating Qt or Gtk dialogs; I
don't know -- if you really want to know, I suggest asking on gnu.emacs.help.
One thing I can guarantee you: if you ask what the module for that is called,
there's an answer you won't get: "Why would there be a module for that?"
Because, it's something Emacs obviously *should* have a module for, and if
it doesn't yet, there's probably someone out there somewhere trying to find
enough time to code one up. So the response you'll get will either be,
"Check out something-or-another-mode" or else "I don't know of anything
like that being released yet; I think so-and-so was talking about it a while
back, but I don't know how far he ever got."

how about

[7-Vodka]

A konqueror plugin that automatically retrieves logins from bugmenot.com.
Man am I sick of being bugged by registration sites.

Re:how about

[dhakbar]

I know you probably do not use Firefox for a good reason, but I must still ask you why you don't use Firefox w/ the bugmenot extension...

Re:how about

[Anonymous Coward]

I'm not the OP, but Firefox is too slow on my system, and Konqueror is nicer anyway in my opinion.

I got one...

[Tumbleweed]

The Internet-enabled 'Remote BitchSlap.' Bad manners, a thing of the past!

I'm jus' sayin'...

My vote:

[BrokenHalo]

An equivalent to Endnote that interfaces with OpenOffice. That would be a big plus for those of us who write scientific documents on linux machines in a MS-based world, where BibTex is not an option.

Oh, and while we're at it, a gui-based stats package along the lines of PASS or SPSS would be nice too.

Re:My vote:

[Geoffreyerffoeg]

On the same subject:

A version of Maxima that's user-friendly, has a reasonable pretty-printing IDE, has Mathematica/Maple compatibility mode, has good online help, etc.

Something that will pretty-print an equation in a given format, either TeX, MathML, or just well-formed math expressions with written functions for sqrt(), etc., and will turn the output into PNG. This would be so nice to hook up to an instant messenger to discuss math problems and such. (Yeah, I'm a math-team nerd.)

Re:My vote:

[Anonymous Coward]

It will come

Re:My vote:

[Anonymous Coward]

Heh, coincidence. I just started thinking about such a project. Since Maxima has TeX output and batch mode, it's not hard to make a GUI where, say, you can type maxima input in one window and get a pretty-printed (pdf, png, etc) equation in another. The tools already exists, it's just a matter of tying them together. I'll put something together and see how it goes.

--
K.

Re:My vote:

[starseeker]

I'm a (minor) member of the Maxima team, and I can tell you a) we intend to do a much better GUI in the future and b) it's going to be a while. Our priority is to fix mathematical bugs in the core system, and modernize the system internally.

Which is not to say you need to wait for the "official" Maxima project to do a GUI. There are several third party programs which do a decent job of being a Maxima GUI. TeXmacs is probably the best place to start, and wxmaxima is coming along nicely.

You can output TeX from maxima, by the way, although if by "pretty-print" you mean line breaking that's a different (and very difficult) problem. The Maxima project will most likely eventually look into doing line breaking the "right way" but non-trivial is a mild description of the issues involved.

Mathematica/Maple compatibility is not really practical, unless you want to reimpliment their subsystems in Maxima. There are too many subtle notational and assumption issues to be worked out for this to be worthwhile, IMHO. Eventually, we might look at a partial "autotranslator" to assist in porting Mathematica/Maple code to Maxima, but it will never be as simple as loading a Mathematica notebook in Maxima. For all but trivial cases, this is too difficult a problem.

Re:My vote:

[sjwaste]

Oh, and while we're at it, a gui-based stats package along the lines of PASS or SPSS would be nice too.

I believe gretl is what you're looking for. It's more like eviews than spss, but it's good software.

Re:My vote:

[Geoffreyerffoeg]

For all but trivial cases, this is too difficult a problem.

It's just trivial cases that I want, actually. I've learned how to type, e.g., Integrate[Sin[x], {x, 0, Pi}] as Mathematica syntax, so I'd like that to be able to translate to Maxima syntax. Rule-based programming, Math'ca-specific stuff like Map[], Apply[], Compile[], and so forth are not what I'm looking for (yet; when I do, I'll learn the Maxima way of doing them).

Groupware

[QuantumRiff]

To really make it take off in business, it needs a groupware system. Not a crazy kludge of different packages, not a web based system. But a groupware server, that supports different clients on different os's. A single administration console, and quick easy setup. all in one little package. Hell, charge $5 a user for it, people will buy it in droves to move off of exchange/outlook.

Software

[hackwrench]

A replacement for the sound system tray icon.
A muliiple desktop, scrolling desktop, desktop in desktop program.
Getting existing open source programs to compile in MSVS.
A P2P, webpage caching, distributed web search database, distributed wiki, distributed any data database program.
A IFS(installable file system) tutorial.
Retrofitting XP features to older versions of Windows.

Re:Software

[Anonymous Coward]

Totally agree on the IFS one! Man, that stuff is a nightmare to figure out.

Re:Software

[nicolas.e]

A replacement for the sound system tray icon.
Too generic to give an answer...

A muliiple desktop,
fvwm and many other WMs. Or do you mean Xinerama ?

scrolling desktop,
X.Org supports this.

, desktop in desktop program.
Xnest ...

Retrofitting XP features to older versions of Windows.
Oh, do you mean on windows ?

Re:Software

[hackwrench]

I amended his question in my mind to mean "on windows" because I believe that is where OSS is needed most in order to transition society from where it stands now to OSS.
When I click on the volume system tray icon it takes forever to open.
I forgot to put it in the list above but a replacement for explorer.exe that is functionally exactly the same but split into separate executables. My notion is that OSS should do a similar embrace and extend approach to Microsoft Windows that Microsoft did in several areas, the end result being that there will eventually be a Windows skeleton that will be easily replaced once and for all with an OSS component.
BTW everybody using Windows XP should download http://taskswitchxp.sourceforge.net/
Does anyone know of other OSS Windows component replacers?
http://www.mythicsoft.com/agentransack/default.asp x is far and away a better desktop search utility than the one that has been bundled with Windows. Problem is, it's not open source.

Re:Software

[Anonymous Coward]

You're a bit fucking slow, eh?

Mapping!

[Geoffreyerffoeg]

I have a Macintosh and I've seen no good modern mapping software for either Mac or for OSS Unix-likes (I don't mind BSD/X Window). In theory, it should be easy to take the USGS TIGER roadmap data set and combine it with local road data from governments and user-submitted manual road additions.

Especially if it's open source, there are some interesting possibilities with it: automatically download USGS's free satellite photos (probably hosted by a certain company's TerraServer), add GPS tracking and maybe automatic road additions, add routing and proper speed-limit data, make a nice 3D perspective view from some point, etc.

Re:Mapping!

[Anonymous Coward]

Not OSS, or even free, but Terrabrowser sounds exactly like what you're looking for. $15, cheap enough, esp. for a Mac program. It's a sweet program, I use it a lot for USGS data, and I just purchased a GPS interface and I'm looking forward to trying it out.

Re:Mapping!

[josecanuc]

We sort of have something like that in the Xastir project.

It true purpose is a client for the APRS (Automatic Packet Reporting System) used by Amateur Radio operators to do object tracking, but as a side-feature, it supports a multitude of map formats.

We're currently partway through integrating GDAL/OGR, which will add a couple dozen new map formats, including TIGER. TIGER support exists if you compile in GDAL/OGR, but it currently doesn't allow customizing of line widths/colors per layer. Fortunately, some users have converted all of the TIGER data to shapefiles, hosted at http://xastir.tamu.edu.

Xastir uses an awk-like language that we call dbfawk for describing the attributes of line segments in shapefiles, which allows you to set the display to look like whatever you like, with some work.

Be default it does GPS tracking -- that's the initial purpose. If you're a Ham radio operator, you can even rig it up to a radio and TNC (radio modem) and see what other hams in your area are up to.

It can also download raster maps on-the-fly from various sources including the TIGER map server, TERRAserver, and maybe another. A "GEO" file describes the raster image source as well as the tiepoints (pixel->real coordinate) for ANY OTHER raster image you have. And the image source can be a local file or a remote file via HTTP -- so getting live radar background to your GPS tracking is a snap.

Items not yet added, but requested: routing, speed limit display, and 3D views -- all those were also requested by our users.

It has extensive documentation and includes explicit instructions on running it on MacOS X.

Re:Mapping!

[baronben]

There is Grass (and if you're looking for an OS X version, there is one).

However, GRASS is about as easy to use as Linux was 8 years ago. Its mostly command line driven, with most of the interesting features (DEM models, buffers, raster calculations) being hidden in either half a dozen menus or obscure CLI commands.

On the other hand, ArcMap, the industry standard, costs tens of thousands of dollars, and is about as easy to use, so you can take your pick.

It's obvious!

[El]

The one thing missing in order for OSS to be on a par with the offerings from Redmond is a really good solitaire program! Currently, Windows is hands-down the best operating system to run if all you want to do all day is sit around playing solitaire! Yes, that's Windows(TM), the choice of computer solitaire players everywhere!

Re:It's obvious!

[Rysc]

On the contrary, PySol is the best solitaire suite, bar none.

Re:It's obvious!

[Darth_Burrito]

If anyone ports Hearts, they need to make sure that, when the computer has a chance to unload a card, he continues to get rid of the Ace or King of spades before the Queen (the bad one). It just wouldn't be the same otherwise.

Good Office Programs....

[metalmaniac1759]

And please don't say use StarOffice/OpenOffice. I've been stuck with them for the past week or so. Everytime either of them starts my machine come down to a crawl! It's frustrating!

Besides that, they *still* dont have a proper Numbering/Bullets subsystem/module. Try doing anything even remotely complex with bullets and you'll end up with freaking results!

On a totally separate note, I was thinking yesterday about how much porn is out there already. Most categories are filled, but I wonder about what kind of porn still needs to be shot. What kind of porn would you like to see which is currenly not available? ;-)

Nandz.

Re:Good Office Programs....

[megabyte405]

AbiWord Word Processor and Gnumeric Spreadsheet. Fast, functional, non-bloated. (Disclosure: I help out with these projects but have used the competition.)

Pr0n Indexer

[Anonymous Coward]

I need a program to index all my pr0n!!!

Voice-to-text, shockwave player

[msouth]

no biggies--can you have that by next year? :)

Re:Voice-to-text, shockwave player

[drinkypoo]

I second the voice to text stuff. We have text to voice, we need the other direction. I want to be able to build linux appliances that take voice commands.

ReactOS

[Geoffreyerffoeg]

The rest of ReactOS. Sure, it's not philosophically pure, but it'll help a long way in OSS adoption to have a Free operating system that's binary-compatible with MS Windows. E.g., in our school system, we've standardized on a particular gradekeeping software, and I cannot forsee the adoption of Linux partially for this reason, that InteGrade won't run on Linux (and WINE seems, by its basic premise of pseudoemulation, too potentially unstable for the administration to trust with grades). Someting that can run InteGrade directly has a much better chance of getting adopted.

Oh, and with MS's free-as-in-beer DirectX drivers, it may possibly play Windows games without rebooting into a spare Windows partition. Eventually, we'd want to reimplement the DirectX API, but ReactOS is at an advantage here as it will be able to run MS DirectX itself.

Re:ReactOS

[Anonymous Coward]

Why not just use WINE? I know you said it's unstable, but ReactOS is not really any further ahead in this regard. If ReactOS was given the full Windows API, you could have put the API into WINE and you'd be just as far ahead, and in a Linux environment.

Reskinner

[Geoffreyerffoeg]

We've got talented graphic designers among us; get them to design a nice window interface, and then implement this one skin for as many display systems as we can. (Use an open interface so another graphic design can plug in easily.) Too often, the display system is unprofessional at best and confusing at worst.

Color management software

[HughsOnFirst]

I won't get into what Photoshop does that GIMP doesn't, but probably the main thing that keeps photographers from considering Linux is the lack of color management software.
I realize that patents are involved, but any field that depends on people agreeing on the way something looks ( separated by space or time ) can't use Linux as it stands.
Sure you can create or edit content on Linux, but if you need a Windows box or a Mac or maybe a SGI or Sun to look at it or print it, what's the point?
I'm on the verge of moving from Windows to Apple for photography, and I'd love to at least consider Linux.
If someone can show me I am full of shit ( about color management on Linux ), I'd be ever so grateful.

here's some

[Anonymous Coward]

Software modem drivers, like what Linuxant does

An interactive disassembler, like IDA Pro

Any good music production software, up to the same standard as, say, Cubase or Reason

Quicken Quicken Quicken

[Free_Trial_Thinking]

We need a quicken clone that is just as functional and just as easy to use.

When I was searching for a way to pay my bills online a lot of banks offer quicken files as the only way to synch your data. I'd love to use some of that Quicken functionality, but Quicken is yucky, yeach!

Spreadsheets

[Free_Trial_Thinking]

Can I please have a spreadsheet program that doesn't limit me to 64,000 rows? If anyone knows of an easy way to do this please let me know! I've been searching for years. Otherwise please build it into gnumeric, OO, etc, please!

Re:Spreadsheets

[magefile]

WTF? If you have 64,000+ rows, doesn't that suggest that gnumeric/OO/Excel are not the right tool to be using?

Re:Spreadsheets

[Free_Trial_Thinking]

Sir, if Excel is the right tool for 64,000 rows, isn't it possible it's also the right tool for 75,000 rows?

I'm not saying 10 million rows here, just that all my datasets are in the 100,000 rows range and my co-workers can only use Excel :-(

Re:Spreadsheets

[magefile]

I'm just wondering what your datasets are and why you're not storing them in a database.

Re:Spreadsheets

[Free_Trial_Thinking]

My job was to do searches through a dataset of about 2 million records for certain characteristics and then put them in csv format for my coworkers to "analyze". They used the filter functions and pivot tables in Excel to do their analyses.

A database may well work for them but they just wouldn't be willing to learn how to work a database or anything much different than Excel.

Plus some things that spreadsheets are good for on a small scale like visually manipulating data and writing functions in cells sometimes are still useful on a larger scale. (with fast scrolling and such).

Re:Spreadsheets

[I_Love_Pocky!]

If your co-workers are stuck with Excel it wont matter what program you are using. They wont be able to deal with the extra rows.

Re:Spreadsheets

[wrt]

It would be useful to have this capability. If you dump several thousand data points off of a piece of test equipment, and want a fast plot, Excel etc is where you'd turn.

Re:Spreadsheets

[kbielefe]

Is increasing the value of SHEET_MAX_ROWS in src/gnumeric.h and recompiling easy enough for you?

Being able to solve problems like this (making a change that almost no one wants, but is a deal-breaker for those that do) is just one reason why I love open source software.

Re:Spreadsheets

[hankwang]

If you dump several thousand data points off of a piece of test equipment, and want a fast plot, Excel etc is where you'd turn.

$ gnuplot
gnuplot> plot 'datafile.txt'
gnuplot> plot 'datafile.txt' usi 1:2, 'datafile.txt' using 1:3

See: Gnuplot homepage. Available for both Linux and Windows.

Re:Spreadsheets

[Matt Perry]

If you have 64,000+ rows, doesn't that suggest that gnumeric/OO/Excel are not the right tool to be using?

They are the right tool as they meet all the requirements except for the arbitrary limitation on the number of rows. If you have a single table of data with a lot of records, why resort to a RDBMS to deal with that? Spreadsheets are perfect for displaying that single table of data and then looking at it in different ways, or editing parts of it quickly and saving back out.

With a RDBMS you'd have to write some code to import the data, then use SQL queries to extract the data and some more code to format it in a way that works for you. Sure, you could put it into an HTML table, but what if you want to lock the top row and the first column like you can in a spreadsheet? Then you'll probably have to end up writing a custom application just to present the data.

Re:Spreadsheets

[Matt Perry]

Is increasing the value of SHEET_MAX_ROWS in src/gnumeric.h and recompiling easy enough for you?

Why is that even necessary? Why the arbitary limitation? Gnumeric shouldn't have a max row limit until the computer is running low on memory.

Re:Spreadsheets

[kbielefe]

I have no idea. You have my vote. The first time I looked at the code was last night to write my reply. My guess is that it was just easier to do it that way.

Where is...

[empirionx3]

any good music notation software?

I know that a TeX combo can do it, but where is a graphical interface?

Re:Where is...

[Tet]

I know that a TeX combo can do it, but where is a graphical interface?

Right here.

clarity

[elliotj]

oss needs clarity. think about the mac. everything is immediately obvious to the end user. apps are sensibly named, things make sense. look at .net on windows. again, this is an easy to understand system.

oss is great, but since it's a voluntary collaboration, things are named after inside jokes or poor conventions. the "k"rap naming of kde stuff for example. or "vi", "gawk", "sed" etc. come on.

I know this will be flamed, but I think some kind of clarity council should be setup to provide consistency and simplicity across applications, tools and platforms. with a bit of this kind of organization, linux could really make a dent on the desktop, and new developers and users wouldn't face such a high barrier to entry.

Re:clarity

[Gilk180]

I agree that clarity comes in handly, but I think the main issues you may have are with the distributors.

I use Fedora (bring on the flames) at home and a Mac at work. Everything I can think of on the Mac is easy to find on my Fedora box. Whatever you have chosen for email is labeled email. web is "Firefox Web Browser". Text editor is "Text Editor".

As for programs like vi, gawk, sed, etc. anyone whoe needs to use or knows how to use any of these as well as grep, lex, bison, emacs knows the commands right off.

Re:clarity

[seanellis]

I agree. On Windows, to install something, I download it and run the installer. On Linux, I either have to use the RPM manager front end (which means telling it where to find the RPM repository, etc.), or I have to use "urpmi" or "apt-get" or some such.

I am a linux user, and I believe that it is superior to Windows in many respects and is getting better, but I do get frustrated finding out how to do stuff because I can never remember the name of the application I need.

On a related note, we need to make things rely less on editing configuration files. IMO, that's just dumb. (I had to do a load of this just to get my scanner working.)

Requiring a user to edit a configuration file with a free-form text editor is just asking for trouble. There's no verification of data values or even basic syntax, error outputs are cryptic, and so on.

Re:clarity

[Just Some Guy]

think about the mac. everything is immediately obvious to the end user. apps are sensibly named, things make sense.

Safari.

Anyone else wanna complain about OSS application names while I still have a fresh set of non-Free counterexamples in my head?

Re:clarity

[drinkypoo]

Let's go surfin', surfin' safari.

Re:clarity

[Grishnakh]

KDE on SuSE is the same way. All the KDE applications, with their krazy K-names, are renamed in the menu to more mundane names describing their actual function.

The solution isn't to rename applications, it's just to make sure the menu describes them well.

Besides, do we really want our computing environment to be as bland and soulless as that of Microsoft's, where every application looks like it was named by someone with absolutely no creativity? If Microsoft made a car, they'd call it the "Microsoft Motor Vehicle", or their truck would be the "Microsoft Cargo Vehicle". Sorry, I'd much rather have a car with a real name. Even the military gives real names to its ships, instead of only numerical designations.

Re:clarity

[evilviper]

with a bit of this kind of organization, linux could really make a dent on the desktop

It's quite funny how every Linux user around knows the ONE THING Linux needs to be the best desktop OS ever, and they're never the same thing. Most of the ideas get implimented after a while, so Linux should be taking over the desktop market any second now...

Wait for it...

Wait for it...

There it i... nope, sorry, jumped the gun. Keep waiting...

Wait...

Wait...

Re:clarity

[r3m0t]

Agreed.

Gnome by default has menus showing the descriptions of programs. "Image Viewer" will launch "Eye of GNOME" (an awful name IMO), "Archive Manager" -> "File Roller", etc.

KDE has a preference to turn this on (or place these descriptions in brackets, or add the name in brackets, etc) but the only way to change this in Gnome is to right-click every item and change the name. I don't mind.

Examine the Gnome HIG and you will see that any distribution changing this behaviour is not following the HIG.

The problem with this (the Gnome way, that is) is that third-party apps begin to hack at the beauty. "Openoffice.org 1.1.2 Writer"; "Gaim Instant Messenger"; "Galeon Web Browser" all go against the grain.

Re:clarity

[r3m0t]

Is it called Safari on all the menus?

iLife.

Please, tell me some more! Gnome apps rule in this. The menu items almost always make sense. (Except for the games and, well, "Archive Manager" is a tad complicated.)

Heres a few:

[comwiz56]

1. Some decent games that can grow in popularity (Yes I know there are issues other than people "just doing it.")
2. A good (cross-platform) Winamp alternative w/ a comparable plugin system
3. Something for retagging and organizing an mp3 collection (possibly even something w/ a database to identify songs against)
4. Anti-virus

Re:Heres a few:

[forsetti]

2. XMMS is pretty decent, and has a handful of plugins
4. CLAMAV, Sophos, OpenAntiVirus

Re:Heres a few:

[comwiz56]

XMMS is decent, but seeing as I primarily use windows... not much use

Re:Heres a few:

[Noksagt]

Read the README.win32 included in the source archive. It can be made to work natively or under cygwin.

Re:Heres a few:

[ted_nugent]

Try ZINF or CoolPlayer.

Re:Heres a few:

[Grishnakh]

1. Well, there's TuxRacer, Pinguins (not sure if that's the right name; it looks like Lemmings, but with penguins instead), etc. But if you're looking for serious alternatives to the big commercial games, forget it; it takes way too much effort and time to create a game of that size. This is one area where I think that commercial software is actually better suited for. What's needed is more acceptance of Linux as a viable market for games, so that the game makers will port their products to it. This has been done before, with Loki, but I think they were a little ahead of their time.

(Before I start a flame war about commercial vs. OSS, I actually prefer open-source games to the commercial ones, because I usually prefer board games like Mahjongg, Chess, etc., which are much better handled by OSS: they don't need fancy artwork, 3D graphics with textures, etc. But if you're looking for a full featured FPS shooter, I think that's just too big a task for an OSS project to stay focussed on. Look at Parsec.)

2) Someone's already mentioned XMMS. Asking this in your list just shows that you're very ignorant of what's out there in the OSS world.

3) There's lots of utilities for this, but they're not so big and well-known as something like XMMS. I use "grip" for ripping CDs and encoding them as MP3s, and automatically naming and tagging them. It uses freedb to get the song names. Try it; it works quite well.

4) Linux doesn't need antivirus software. This is just a byproduct of Windows' terrible design; since Linux doesn't have the same shortcomings, a band-aid solution like that isn't necessary.

Now if you're talking about spam-filtering software, that already exists: spamassassin, etc. But the only reason to use anti-virus software in Linux is because your Linux box is a Samba file server for a bunch of Windows clients, and you want AV on the server. And again, there are products for this as another poster has pointed out.

Ay See Pee Aye

[Halcyonandon]

ACPI support so that hibernation is reliable, and doesn't take 5 minutes, and so that suspend doesn't drain the battery as quickly as just leaving the laptop on! Without applying 597 patches...

Vector Drawing

[poindextrose]

I'm rather impressed with Sodipodi. Not exactly an Illustrator killer, but good nonetheless.

Re:Vector Drawing

[Drantin]

I know it was based on the same codebase, but how does inkscape compare?

Educational software

[LoveMe2Times]

Last week I was poking around the web looking for good music theory/ear training software. Seems like a perfect thing for OSS, and in fact GNU Solfeg exists. The non-OSS competition is lots of shareware and crapware. But finding the OSS program was hard and the crapware was abundant.

This kind of stuff is perfect for OSS: it's pretty straightforward (no real innovation required), tends to be loosely coupled, benefits immensely from depth/thoroughness, and the competition doesn't tend to be super-slick. It's lots of the same reasons that WikiPedia is so successful.

Re:Educational software

[Anonymous Coward]

That's excellent, thanks for the pointer.

Heres something...

[spreerpg]

A Mathematica replacement, please.

Re:Heres something...

[NotoriousQ]

Mathematica is available on Linux.
There is also Octave for the Maple fans.

Re:Heres something...

[Noksagt]

Mathematica is under active development by some smart people. It will be difficult to catch-up. Still, Maxima is really quite good. Many of the Matlab clones (scilab, octave) are also great.

Re:Heres something...

[John Meacham]

Maxima (maxima.sf.net) is a great computer algebra system. Highly recommended.

Re:Heres something...

[suso]

Catch up? As one of my sig lines says:

5 years will not be a long time to wait 4 years 364 days 11 hours and 59 minutes from now.

In other words, get started now and in five years you'll have something respectable. Who cares whether you've "caught up". Gimp may not have caught up with Photoshop, but its damn spiffy.

Re:Heres something...

[Noksagt]

I do agree & consider Maxima spiffy. Several years ago, it wasn't GPLed & was one of the leading mathematical packages. When I suggest it to people who are used to Mathematica, they alway tell me of different features Maxima lacks. So I merely meant to provide a disclaimer.

Re:Heres something...

[ktulu1115]

Never used Mathematica, and not sure if they're related, but a MathCAD and MatLab replacement would be great. Havn't tested their comptability under Wine yet, it'd be interesting to see if they worked.

Speech Recognition!

[zeno53]

Those of us who must rely on speech recognition software to use our computers not only have no choice but commercial software, we are also relegated to Windows or, with arguably rather less function and accuracy, Mac. There is one open source project of which I'm aware, CMU Sphinx, but its progress is slow and it is not usable as a complete, continuous-speech, large vocabulary speech recognition solution. NaturallySpeaking for Windows is an excellent product and allows those of us with disabilities who cannot use keyboards to do what you see here and many of us who have paid for the software over the years don't begrudge supporting the company but, given its product is not available for any platform but Windows I think this is an excellent opportunity for OSS.

There are others who feel the same. In fact, for a time, IBM had a version of its speech recognition engine available for Linux users but it has since abandoned the project. So, OSS would seem to be the way to go, perhaps building on the very promising CMU Sphinx.

[Dictated using speech recognition technology. There may be air oars]

Re:Speech Recognition!

[Anonymous Coward]

given its product is not available for any platform but Windows I think this is an excellent opportunity for OSS

Let me second this comment. I am an utter nonprogrammer (HTML, a little JavaScript picked up on my own, is all). About the only technical credentials I have involve case modding, so I'm kind of an outsider here. However, as a scholar in the humanities I have used Dragon for years to write up my work--and I'm looking for a change. Now is definitely the time for an open-source continuous-speech dictation app.

I have gotten more and more frustrated with Microsoft's continuing restrictions and do-it-our-way attitude, and the only thing keeping me from moving to Linux--despite whatever technical difficulties I might encounter, despite the fact that I might have to use up vacation time to learn enough, etc.--is the lack of a viable speech-recognition program. That's an absolute must; I just refuse to type anymore. It wouldn't matter if I had to pay for it. I've gotta have it, one way or another.

And that's getting to be a major problem. The latest release of Naturally Speaking, version 8, requires that you give "the company" (as zeno52 calls them) your computer image over the web, which they then use to limit you to 6 installs on different machines over the life of the program. And they define a "different machine" as any machine that's undergone a hard-drive or motherboard replacement as well as a CPU change. For someone like me, who messes around with CPUs, peripherals, and hard drives, and has to reformat every once in a while when an experiment goes wrong, the new installation policy isn't just another arrogant sign of encroaching DRM--it simply won't work. I have no choice but to forego the upgrade.

But the new policy shouldn't surprise anyone who knows the financial ins and outs of "the company". About four years ago, Dragon Systems sold out to Lernout & Hauspie, who then folded and were acquired by ScanSoft, whose "last name" ("-Soft") is no accident. Through various financial channels, M$ (or others tied more directly to M$) has a major stake in "the company." It's basically an extension of Redmond.

As soon as there's a working Linux dictation app, I'm kissing Microscum goodbye forever (even on my Windoze box, the only M$ software I have is Win98 itself--I even use 98Lite to rip out IE). But that "as soon as" is a big one. Please hurry, guys! If an open-source VR dictation app were close to completion, I'd happily write the user's manual, but that's about all I could do. The rest of us need you! If any of you programmers are interested, I located one other OSS effort aside from Sphinx: http://freespeech.sourceforge.net/FreeSpeech/html/ seems to have stalled but could undoubtedly be rejuvenated with enthusiastic new contributors.

Digital A/V

[Apreche]

Digital Audio and Video editing. Stuff like Premiere. Acid Music. Reason. CoolEdit. ProTools. iMovie. Doing audio and video editing should be as good on Linux as it is on a Mac.

Also, burning of optical disks. Yes k3b exists. yes, technically you can burn just about anything. But nothing linux can do comes close to Nero. I need all the perfect functionality of Nero in linux.

Steam. The only pc game I play other than puzzle pirates, which is java, needs to run better on linux. Using cedega I can only get the resolution up to 800x600. Anything higher drops the framerate from perfect to less than 1 fps.

Someone else mentioned audio mixers. Alsa is very good, and is about as good as I can expect, but not as good as I can hope for. I have an SBLive! Value with the latest alsa kernel drivers. It works and plays music very well. But if I use winamp in windows with directaudio not only does the mixer work properly and is labeled correctly, but the sound quality is imporoved tenfold. I don't know what the difference is, but even my non-audiophile self can hear a noticeable difference. All my friends hear it too. It's the same hardware, it should work the same regardless of OS.

That's really what linux has to do now. Firefox and 2.6 brought us to the top of the hill, we're just nearing the peak of the mountain. We have support for most hardware and enough software to replace windows as a non-gaming desktop machine that is technologically superior in almost every fashion. But a lot of the hardware support is existent but non-perfect. Sound works, but not perfectly. CD burning works, but not as simple and perfect and beautiful as nero. ATI cards work, but are a pain in the ass. Nvidia cards work, but with closed source drivers. It's like everything works 90% perfectly, we need to push it to perfect, then linux will have smooth seas.

Re:Digital A/V

[DaveJay]

Hear, hear. Reason and Cakewalk Sonar are two programs I use on a daily basis.

So those, and a good software/driver installation program that works using a standard interface. Yes, I -can- manually install and compile when necessary, and yes, I -can- roll my own kernels and whatnot, and I do. But I really really really really really would love to have a program like InstallShield or WISE to make it easier. Dreaming...

Personal Finance Manager

[Noksagt]

GNUCash is great, but has a lot of dependencies & some aren't used to double-entry. We need a simpler Money/Quicken clone. Checkbook Tracker made a fast start on something like this, but we don't really have a Money-killer the way we have an Office-killer.

Re:Personal Finance Manager

[megabyte405]

Google for Divifund in a while (new name, website, release) and you'll find a personal budgeting tool for Linux (*nix - Python and GTK+) and Windows.

Disclosure: I help with the development and marketing.

Re:Personal Finance Manager

[Skater]

...and GnuCash is about 2 gnome versions behind. Very frustrating to install on Slackware (see my website).

--RJ

EDI

[redog]

Preferably written in python :-P
AS2(this is how walmart places orders)

Darn Few

[rueger]

It actually came as a surprise when I looked over my desktop and menu to see what I might add to the list. I can't find much on my PC that isn't already OSS, freeware, or shareware.

Sure I'd like an OSS mail client as good as Pegasus, but it' still free so that's not an issue for me.

Beyond that the only decidedly non OSS/free package that I use a lot is Dremaweaver. Oh, and Palm software of course.

Honestly, of the major apps that I rely on day in and day out, almost nothing is store bought.

And that's a pretty good sign.

Re:Automated

[Anonymous Coward]

Easy... run wget as a crawler from your favorite launch portal (try persiankitty.com).

Trust me, you don't have enough drive space or bandwidth to "finish" the download.

Office

[Wade Tregaskis]

It's that simple really. OpenOffice/NeoOffice/whatever are not suitable replacements for MS Office on a Mac - and this "but Firefox is popular and it has a shit lowest-common-denominator interface" is not helping.

A module to interface Evolution with Exchange...

[Kevin Burtch]

Before you reply with a knee-jerk reaction telling me about the Evolution Connector (originally by Ximian), it is useless unless your company runs OWA (Outlook Web Access).

What we need is a module similar to Connector that works via MAPI, not HTTP, for the majority of companies out there who do not run a web server for email.

I would _love_ to be able to run Linux on my main work computer, but unfortunately I have to stick with MS-Windows for that system to be able to communicate with the rest of the company, and run Linux on an abandoned dinosaur for real work.

Security holes!

[Lord Bitman]

and by "security holes" I mean "more ways of letting one program communicate with another program", or more specifically, "less ways of letting one program communicate with another program, but more often"

ClickBook

[Donald R. Weimann]

I would like to see a program that works like ClickBook from Blue Squirrel. It is a program that will take anything you want to print and turn it into a booklet, flyer etc. I for one would gladly pay for it. That is the only reason for which I have win4lin on my computer.

Re:Automated

[Anonymous Coward]

Yesssss... nget... Drink deeply from the firehose of pron that is usenet!

Mod troll down

[Anonymous Coward]

Which did you use: Star or Open? On which platform? Which version? What problems do you have with bullets/numbering. Either has been excellent replacements for quite a while.

GIS

[catfoo]

a full GIS platform that is ussable by new gis users and still usefull and customizeable for expert users. a simple mapping application that makes it easy to deal with multiple datasets, make maps, and do basic GIS queries yet can be customized into verticle applications. THUBAN and QGIS (and the libraries they leverage) are two efforts that are approaching this but they are a long way off from filling the need. such a GIS application doesnt need to completely replace the the high end applications developed by ESRI (and others) for those will allways be used by GIS techs but such a app would make GIS more accessable at occasional user end.

Technology isn't the problem...

[Zarf]

business deals are the problem. You need to make the right business deals with the right people. Get on that okay?

PhotoShop killer (GIMP)

[JMandingo]

GIMP is an OSS shining star, and a marvel to behold, but GIMP still has a way to go to be much more than a toy for those who do professional grade graphics. Every year I try it out to see how it has advanced, and every year I find myself going back to PhotoShop because some feature I require is not present yet. Maybe next year.

Re:PhotoShop killer (GIMP)

[jimicus]

Every year I try it out to see how it has advanced, and every year I find myself going back to PhotoShop because some feature I require is not present yet. Maybe next year.

I keep on hearing comments like this, on Slashdot and on various mailing lists. I cannot help but wonder how many people who would like to see Gimp become a true Photoshop killer, try it, find a feature that's missing and DON'T submit an enhancement request.

This isn't commercial software! New features do not automagically appear! Even if you don't have the expertise to write it yourself, submitting an enhancement suggestion shows what features are in demand.

Re:PhotoShop killer (GIMP)

[meme_police]

Lack of features aren't why I stopped using the GIMP. Hideously slow performance is.

Games

[j1m 5n0w]

Given how many people dream about writing games, it's surprising how few good open source games there are. Perhaps what's lacking is a good framework - few have the time and abiblity to implement a whole high-quality game from scratch.

I'm quite impressed with stratagus, though. It seems like a reasonably hackable RTS game framework.

-jim

A translator... and a controller

[zogger]

...no, not urdu to engrish or klingon, I mean GUI to CLI then to AUI. I'd really like to see a program that would let me see what any random button mashing or input in a GUI interface *does* in the form of a normal scripting command and also to see the hierarchial tree of files accessed in real time. It can be "focus" based on the other apps. This is a learning tool. An exploration tool. A linux useability and make-more-practical tool.

And I'll second the request from down the thread, a speech to text and text to speech, eventually leading to speech to speech,diggit, the dang talking computar. Audio User Interface. You get one of them babies, you'll be set, in like flint, leet, topdog. People freeking talk to each other, they DON'T stand next to each other and type at each other. We are audio visual creatures, tactile is down the list of senses. And don't forget the aging of the population and how arthritis and whatnot screws up your typing ability, let alone how it even affects younger folks who do it a lot.. Big ole useability hint there to anyone looking for an actual folding money market of some kind.

I used to have a little mac classic proggie, forget the name, but actually worked well, you could request apps on and off, etc, verbally. Something like that, but *more*. First, the speech to text, because THEN not only is it handy for those with disabilities who want to compute, but you could use the text output to run the computer.

Lotus Improv like next generation spreadsheet

[John Meacham]

Lotus Improv.

We are already doing a good job of replacing and improving all the software that made it. How about taking some good ideas from the past which didn't quite make it in the commercial space and giving them new life as OSS?

Re:Database,

[tengu1sd]

Code Generation and other one shot down, dirty, fast and ugly projects. The last database conversion I was tasked with at my old company, I used Access/VB to track specs, create customer reports and define the job. Drove the project manager crazy, he kept asking how many lines of code have you written as if that had any bearing on the project. After wasting a month talking to the customer, I came in on a Friday, slapped some VB around the database and generated just over 17,000 lines of code. Then I wrote a few hundred lines to push data, ftp'd the whole thing to a real system and went to lunch. Project closed 1 week early, customer knew exactly what I was doing, no cleanup.

Access works for one shots. It's the PHB that wants a multi user ERP, customer management, HR application, inventory ordering and work flow management form that gives it a bad name.

OSS

[Deliveranc3]

Needs a business model that works.

And someone to demonstrate it.

I know it sounds ugly but there are lots of projects which can't be developed under opensouce because there is no room for liscencing costs etc, if there was a plausible businessmodel which generated even limited revenue they could use that money to smooth some of the area's that only money can smooth.

Such as liscensing fees.

It's been mentioned before,

[MoOsEb0y]

but I would really like to emphasize this. Linux needs decent quality DVD authoring (read GUI... not dvdauthor) and transcoding software. K3b is probably the easiest place to integrate transcoding (they already have support for DVD -> DivX), but I am talking about MPEG2 -> MPEG2 to make dual layer DVDs fit on 4.7 Gigs. This is the place where I feel that linux is lacking the most. That and ATI's drivers suck, but that's not our fault, right? ;)

Re:It's been mentioned before,

[Gleng]

That and ATI's drivers suck, but that's not our fault, right? ;)

Yes, they do :)

But! The open source "radeon" driver has been vastly improved in xorg 6.8.1, and seems to work at least as well as the ATI binary drivers on my Radeon 9200.

It's definitely worth checking out if you're having no luck with ATI's releases.

Re:It's been mentioned before,

[korgull]

qdvdauthor worked well for some simple DVD's I've created. It needs some work but it's a good start.

License management tools

[wayne606]

A way to manage licensed software, using nodelocked or floating licenses.

Products like FlexLM cost a lot of money. But you release a version of your product and then the next day somebody in Bulgaria is selling cracked copies for $50. I'm not saying that an OSS product could do better, because a clever hacker can work around any type of protections... But if you're getting a leaky boat at least you'd rather get it for free...

It may seem strange to produce a OSS license manager, designed only to help others make money off non-OSS products. That's probably why nobody's done it ... Also once enough products use something like FlexLM, the customers' IT people get used to it and consolidate all their licenses on a single server, and if you give them something new and strange they don't like it.

Re:Database,

[Anonymous Coward]

It's the PHB that wants a multi user ERP, customer management, HR application, inventory ordering and work flow management form that gives it a bad name.

That's the problem, that's all I've ever seen it used for.

Clippy...!

[bergeron76]

and/or Bob would be nice.

Better still, how about some games?

Even better than better still, how about Business Plan Pro Premier?

some ideas

[Darth_Burrito]

I'm sure some of these things may already exist....

Some newish application ideas....

* A Web Browser that can be used by the blind and by sighted developers. Something like JAWs for windows would be even better.

* An open source php ide that lets you debug things as easily with php/apache as you can with .net/iis.

* Access-esque database software. Eg, something that can serve as a nice front end for just about anything else.

Suggestions for improvements...

* A standard set of uniform, good, easy to use, fully featured, graphical configuration tools for popular software like apache, mysql, postfix, etc. Think redhat-config-blah but works on every distribution. And not just some graphical front end that exactly mirrors the file format (like rh's samba config), but actually something designed to help make configuring things easy. Not all wizards are evil.

* Continued improvements to the GIMP. The recent version is soooo much better, but sometimes it still makes me cry.

* phpmyadmin is functionally incredibly awesome, but desparately needs a graphics designer or ui engineer or both.

* If I had a dollar for every time I had a broken package that couldn't install itself... Why not have apt or whatever automatically log problems with package dependencies to a central location, perhaps even give you a link to a bugtracker issue or forum in which people could help each other out.

* Friendly distribution specific idiot proof installation packages for all those projects I want to try out but either can't seem to get to work or haven't had the time to bang my head against wall yet: mythtv, zoneminder, freevo, etc.

* Documentation. No wait. Not just documentation. Mentally scalable help starting from inside the application and spreading outward. In other words, it would be nice if, upon encountering something unknown, you could maybe hover over it and get some alt text. If you're still confused, maybe there would be an option to click on an inconspicuous help link that would bring you to more detailed information on the immediate topic, and then some place to go from there to get a broader view. Integrated scalable help.

Simple graphics editor

[ted_nugent]

Going from TuxPaint to the Gimp is literally like jumping between kindergarten and college. Is there nothing inbetween for simple drawing and photo touch up?

Re:Simple graphics editor

[quamaretto]

I certainly feel that anything with the word "paint" appended is probably underpowered for the job. On the other hand, I don't feel Gimp is an overpowering tool even for simple photo retouching.
Gimp is actually a fairly compact tool compared to Photoshop, which insists on being in a huge window, takes much longer to start up, and is designed for images with 100's of layers and a high speed filter for absolutely everything.
The trouble with Gimp, to my way of thinking, is that it's just a bit rough to find the right tool and dialogue, which was improved from 1.x to 2.0, and will probably improve more. YMMV.

Re:Simple graphics editor

[r3m0t]

Gimp is actually a fairly compact tool compared to Photoshop, which insists on being in a huge window...

As opposed to GIMP, whose slew of windows is extremely confusing and requires its own workspace, which stops me from seeing all my other applications. Just like Photoshop.

If only GIMP had MDI, I would be... happier with it than I already am. I already am happy.

I understand it's some Gtk+ problem stopping them?

Re:Simple graphics editor

[quamaretto]

I admit Gimp's windows are confusing at first, if you make no attempts to change them. Personally, I can make do with putting all of the dialogues in one bar and not using the screen space, or not having most of them visible until I need them. Also, I really don't like MDI just because I find it clumsy. If Gimp had a tabbed setup like Firefox or something else with less metaphor shear, I would use that.

Nitch business software

[mikefe]

Tax software

Personal Finance (need to look at GNUcash again, but isn't there anything else?)

Also when I was shoping around for a used car with a friend a month ago, I saw a bunch of interest calculators / C(ustomer)RM software that would be a requirement on the one or two desktops at a small dealership.

All of these little niches already have software for windows or even dos systems. You can't reasonably ask them to switch until there is a viable replacement for that. And being cross-platform is a strict requirement since the software will need to integrate windows only shops during the time when there is too much inertia against moving to an OSS OS.

Adobe AFTEREFFECTS

[Anonymous Coward]

A good, solid AfterEffects replacement. Sorry, but for professional (read: standards, HD) video editing, nothing on Linux comes close.

Even something as good as FCP would be awesome. Currently the choice is very slim.

DVD authoring!

[Frodo420024]

Something easy that I can use to whip out my holiday movies in a few hours. Ulead MovieFactory comes to mind, or their DVD Workshop for more complex projects.

DVD authoring with OSS still takes a computer science degree and LOTS of time to get right.

All of them.

[Mark Hood]

I know, you'll mod me down as flamebait...

But can we please finish some of the existing ones? Bring them up to a polished standard and people will want to use them, and that'll help the migration to Linux, give us the critical mass of users required to get investment in the next 'big missing application'.

Look at FireFox - now at 1.0, fully capable, (Slashdot rendering bugs aside) and the mainstream media is loving it.

OpenOffice.Org and The Gimp aside, there's not many OSS products that are to the same quality - even fewer if you include documentation! And please don't just list a dozen apps as a rebuttal - look how much developer time is being spent on SourceForge applications in Alpha/Beta/PreDesign status. (Note I didn't say wasted...)

OK, I know people can't migrate their whole office until they have replacements for Access, Outlook, Notes, etc - but if we make all those things 'just good enough' for the geek crowd, the average user won't want to touch them. Believe it or not, people use Clippy to get help in Word & Outlook - I've lost track of the number of OSS projects where the documentation is a hastily-prepared FAQ and the help button opens a window saying 'coming soon'.

Get the leading OSS projects to 1.0 (Or even 1.1 :) and then worry about adding more. No IT manager worth his salt will risk rolling out Version 0.75a of anything unless he's got the support of the MD/CEO - if it fails, even if the MS equivalent would have more problems, he'll be for the high jump for installing 'test versions'.

And document, document, document...

Mark, pouring petrol on his Karma and handing you a match...

Re:A module to interface Evolution with Exchange..

[bonezed]

yes it would be nice not to have to run Outlook

and while my work has IMAP access I still don't get address books etc

A photo-manipulation / paint package

[EnglishTim]

... that doesn't suck.

Yes, the GIMP sucks.

Scientific Graphing

[timbos]

What I'd love is something that works like Origin, Sigmaplot, Tecplot: Graphical Interface, Curve Fitting, Data Analysis, Peak Picking, Data Manipulation etc... I haven't found anything like this, despite a bit of googling around and an exploration of the ports/packages trees in a few distros. This is what is keeping *nix off my work desktop (oh, and proper interfacing with the Novell Network (ncputils causes a kernel panic))

Chacun a son gout

[trewornan]

I agree with you, in as much as, I find it easier to just cut straight to the HTML but, I also understand that some people might prefer the WYSIWYG route - that's fair enough if it works for them.

I tried LaTeX, liked the idea, understood the advantages, worked hard at learning some of it's subtleties but in the end it was just easier for me to go back to WYSIWYG - it suits me better.

Try to be less judgemental (lest ye be judged).

PS sorry can't be bothered to look up accented characters for the subject line.

Air-Marine nav program please!

[deadweight]

Right now there is NOTHING for marine navigation. See http://www.sping.com/seaclear/ and www.fugawi.com for Windows versions of these programs.

education software

[jmayes]

reading & spelling for little kids
math, history, science for school age kids

print shop

[cryptozoologist]

i have a windows 98 partition and the only reason i haven't wiped it out yet is print shop. yes, one theoretically has the functionality of print shop in the gimp, but it is great to pick your paper, pick your layout add some text and graphics and go.

Cinelerra's UI

[Anonymous Coward]

Having recently produced a DVD using DV camera + kino + cinelerra + transcode, I can say that DVD production on Linux is possible. I'd even say that the final result is better than on Windows, since I was able to de-interlace the video.

However, for all its power, cinelerra has the most horrible UI I have *ever* seen! If someone would like to fix it (even just changing the widgets would be a start), I'd be grateful :-)

Re:Cinelerra's UI

[vaeder]

you should try the unofficial cvs version. It comes with an alternative theme. http://cvs.cinelerra.org/

Needs

[michaelbuddy]

Need Video editing.
Need DVD authoring
Need SVG based animation package that exports to SWF as well.
Need to cut the bloat of open office.
Need Multimedia Authoring ((like Director)This will help get you education games, kids games that people want.)
Need Protools killer.
Need NVU to get up to standards and stop changing my tags.
Need XUL IDE
Need Java IDE that is easy as Director.
Need Inkscape to smooth out it's bitmap export
Need to work on Blender Game Dev, make it super easy.

Tame puppy dBASE clone

[Anonymous Coward]

Yeah! I want a dBASE clone. These days even well-informed geeks don't know what it is/have never heard of it.

The file format, the command prompt, the programmability.

dBASE! What a concept! Used to be king of the hill.

Depends

[boodaman]

For corp/business: server-side alternative to Exchange, and client-side alternative to Outlook's calendaring and scheduling. With a simple conversion path, so whatever was on the server could read and utilize existing Exchange datastores without a hiccup. SuSE's product is the only one out there that I know of.

For personal: an alternative to Macromedia's Flash and Director, an alternative to Quicken (besides GnuCash), and an alternative to SonicFoundry ACID (now Sony ACID). With native support for all of the above.

Additionally, some sort of third-party testing/verification system. In other words, a process by which something like GnuCash can be absolutely verified as compatible with Quicken. Yes, GnuCash is open and I can read the source for myself, my point is that to get users to switch, you have to make it easy. If I could go to a site I trusted and see an article that tells me GnuCash supports and doesn't munge my Quicken data, and that switching is just a matter of one or two simple steps, I'd switch.

The more momentum, the harder it is to switch, even knowing the benefits. I know Linux is "better", for example, and that Linux has audio apps I can use, but I have over 300 ACID project files...I'm not going to switch unless I know I can switch to something that can pick up right where I left off with a minimum of fuss. It isn't worth the time to spend 3 weekends, for example, messing around with conversions etc. when I could use that time to be creating.

Almost forgot: a reliable, absolutely stable disk partition manager that would let me resize my Windows partition on the fly so I could install Linux dual-boot. If someone buys a new computer, and they're not savvy, the thought of reformatting, reinstalling Windows in a smaller partition (assuming the OEM install disk they get even lets me do that) and then installing Linux isn't going to happen.

Exchange

[Edgetho007]

Most of my corporate clients would switch today if there were a Exchange server. I would argue that Thunderbird and "Mozilla Post office" would be the lethal combo that could do the trick.

Gnucash

[PigeonGB]

Gnucash is not available for Windows AFAIK and it still doesn't close accounts at the end of a business cycle properly (you have to work around that currently) but it otherwises interoperates with Quicken and Money.

Cards!!

[harrkev]

Wy wife would never accept any OS that did not have some sort of greeting card/sign/banner program. Think Print Master/Print Shop/Greeting Workshop/etc.

I suppose that I could use some combination of GIMP/Open Office to make do in a pinch, but my wife would never go through this trouble.

Also, it will be nice when Linux has enough mass to make GPS manufacturers (Garmin, Magellan) port their map-transfer software to Linux.

Music notation software

[Pseudonym]

We have a lot of MIDI sequencers, but no decent music notation software. No, MusicTeX doesn't count.

My Kingdom for an IDE!!!

[Anonymous Coward]

We need modern, full featured C/C++ IDE oh so badly.

I want completion, refactoring, a class browser, full gcc error parsing (I want to click on a build error and go right to the corresponding source line), a real gui debugging interface (a gdb command prompt won't do), source control integration, and a packing tool to auto-generate RPMS and Gentoo ebuild packages.

It should "just work" with autoconf, automake, libtool, and pkg-config. I want to be able to check a box next to gtkmm or any other properly installed library and have all of the build stuff taken care of for me. Compiler flags, include paths (in the right order!!!), linker flags, library paths, and libraries (again, in the right order!!!).

Eclipse with the CDT works but doesn't integrate well with the Linux way (autoconf / automake / libtool / pkg-config). The Eclipse CDT also borks up it's own makefiles from time to time requiring one to start a new project, copy a all of the build options manually, and import the source files. Besides, Eclipse is java. I like java but this isn't it's domain. Eclipse is laggy on an Athlon XP 2400 and eats up the megs like candy. Why can't we compile it with GCJ yet?

Anjuta supports the Linux stuff (automake, etc.) somewhat well but it's so rife with bugs as to be unusable IMHO. What sucks is that it's just stable enough to lull one into trying it for serious work. Then two weeks into the project it drops the hammer...

I tried 1/2 a dozen other OSS IDE's and found Eclipse /w CDT and Anjuta to be the best out there. Did I miss one? Please tell me I missed one.

I'm stuck using Eclipse for now I suppose. Time to build that eight way 64FX 55 box...

CAD / CAM

[mindaktiviti]

An Open Source CAD / CAM solution would definitely be awesome. Giving technology into the hands of primary & secondary industry companies will inevitably end in cheaper consumer products.

Configuration utility

[yamla]

A complete, comprehensive graphical configuration utility. Something like the xDrake tools provided with Mandrake, but that handles configuration for ALL the software included in a Linux distribution.

Note that such a tool isn't available in Windows. But in Linux, you are expected to configure much of your software using text files. I think this is great, but I wish there was also some nice graphical configuration tool.

There are also some web-based tools such as webmin. They get close, but still leave a LOT of programs to be hand-configured. On Windows, the Control Panel is also close but almost all Windows apps that you install won't stick a configuration widget in there.

Mathematica

[crbowman]

This is an awesomely usefull piece of software and incredibly well done and complete.

OT: sig

[http]

10b||~10b
0x3 ?

Re:OT: sig

[forsetti]

Nope -- think "Binary/Boolean" and "Shakespeare"...

"$MySig....that is the question!"

Capture streaming web videos

[zloof]

I've been looking for a while for a program that will authenticate to a traffic cammera server that broadcasts over the internet, and capture the streaming video into a video file (*.avi, *.mpg, *.ogm... something) I haven't found one that does the job. Several programs will capture streaming web content, but none seem to be able to send a username & password to enable such a capture. It seems like this type of program might be good for a home security system that you want to capture 24 hours of video at a time, and then overwrite that in a continual loop.

I have used some of this proprietary software for parking garages, and traffic signals (www.axis.com) but that software doesn't do as good of a job as I would like. For example, if you are looking at the raw capture data, the picture is smooth, and the frames are crisp. Then you use there "export" function and no matter the speed of the computer (2.6Ghz 512Mb) the frames are not evenly timed and become clumpy and individual frames are blurred.

Bassically what I want is a program like VirtualDub for streaming content. Has anyone seen any programs that might do authenticated streaming video capture?