Ubuntu Linux vs. Windows Vista

[Renegade]

...All new software for me must have Linux versions now or I won't purchase it from this point forward...

-zridling (May 02, 2007, 08:19 AM)

This is where I think that you're going to get screwed. There are very few commercial developers willing to invest in Linux versions.

(I'll spare the commercial vs. free rant for the time being.)

When Linux gets a better adoption rate, then the commercial guys will follow. But for now, that's not happening. It's only a select few using *nix. Well, OSX is *nix but that's BSD. And even then, it's hard to get a lot of people to develop OSX versions of software.

The problem comes down to the economics. (Ooooo... Evil! Putting food on the table, clothes on the back, and perhaps a few bucks in the pocket... Scary!)

While some things can be done for free easily, or free by someone with the time (at the moment), the FOSS thing just isn't sustainable for forward looking applications. That's always commercial and always has been. There are no FOSS applications out there that approach their commercial versions. (Don't mention Apache - there's iPlanet out there which blows it away.)

What we really need is a fundamental change in mentality. Free is great sometimes, but other times we need commercial software. They are just different circumstances. One isn't better than the other (in general), but they need to co-exist. The FOSS community has been extremely hostile to the commercial sector, and that needs to at least be curtailed or moderated.

If FOSS becomes the dominant model, innovation will stop. Or be severely slowed.

I would LOVE to be able to develop for Linux and OSX, but at the moment, that's just not possible. I wish it were. But the bread and butter is with Windows for end-user applications.

I know that I'm not going to be popular for this here. I'm going to get some flak for it. It's not my fault. That's just how things are. It's the reality of the world that we live in. I'm not trying to say how things should be. I'm only trying to describe how things really *are*. 

For me, I make my living off of *free* software. I'm all in favor of things being free as much as possible. But at the end of the day, someone needs to pay the rent, the electricity, pay for the food on the table. Software takes real work, and saying that it needs to be free is offensive.

Sure - there are some good things there, but...

I don't mean to go off on you Zaine. But there has been a very wrong under-current here about FOSS that I just can't ignore. Somebody here needs to be the bastard that points out the reality of the situation. I just suppose that the bastard is me for the time being...

[zridling]

[renegade]: If FOSS becomes the dominant model, innovation will stop. Or be severely slowed.

That's okay, Renegade, I appreciate your response because you've clearly thought this out. But on this point I think you have a fundamental misunderstanding of what more FOSS will bring to our collective table. I don't believe all software should be free. It's just that if I'm going to migrate to Linux over time, it wastes my money to invest in commercial software that I won't be using in the future, unless that same company ports it to Linux. Not all software on Linux machines is free or open source. And sure I could use WINE or VMWare to run Windows software. But the difference is, I don't want to. The key to your development dilemma is Open Standards, which:

• Promote interoperability among products made by different vendors and software providers.
  
• Drive competition in the marketplace thereby increasing product innovation and quality while lowering prices.
  
• Provide customers with a greater choice of applications and providers.
  
• Level the playing field, giving no clear advantage to any player unless they happen to provide a superior product at a lower cost, regardless of their current marketshare.

Clearly no one would possibly want any of these things, right? If the first point is achieved, then the opportunities for you to make the rent money increase exponentionally. No one more than me wants programmers to go to the bank; because as an end user, I benefit directly from your hard work. You said it best: "Free is great sometimes, but other times we need commercial software." I still say one of the most creative licensing combos is by Don Lessau (XYplorer dev) who sells a cheap Lifetime license, but then leaves a Donation link up on his site. I use it, too, because I know he doesn't make a killing off his software, and because he's on my side. Sure, the vast majority won't donate, and he's the victim of piracy like everyone else. But it's a unique 1-2 combo I haven't seen elsewhere.

But that's also where we have to work to change the mentality toward donationware, for instance. I've written to FileForum no less than a half dozen times asking them to add a 'Donationware' category to their site for devs. Still nothing, and you get a lot of 'reviewers' trashing good software because the dev asks for a donation — any donation. How many times has someone online written: "If only I had a dollar for every time my software was downloaded, I'd be rich." Unless it's highly popular software, not all downloaders will keep it or use it. But if it's something like uTorrent or CCleaner or what have you, you can bet most are using it.

[Renegade]

I'm all for open standards. It makes everyone's life easier all the way around... Well... Except for the monopolies.

The problem you mention with Don Lessau permeates the industry. Freeware really is a good thing a lot of the time, but it makes it difficult also for others to get into the industry and creates those barriers. That's not necessarily a bad thing because it also raises the bar for commercial alternatives as well giving end-users a higher level of quality too. But for those where they do offer a good product, freeware reduces the value making it difficult for those other good products to get to market or to stay in the market.

Let me be perfectly explicit there - Most users do not understand the difference between quality software and complete crap. A plethora of freeware crap lowers the overall value of the market. e.g. There are some VERY high quality MP3 players out there, but this is NOT a good market to be in because there are a TONNE of crap players as well and everyone has been forced to give things away for free. The same sort of thing happens in other markets.

The flip side of the coin is that the really good stuff that competes with crap needs to have a higher price point, which drives prices up and up. Want some really solid web analysis software with decent licensing conditions? Great. That'll be $14,000 or so. (I am not joking.) This is NOT good for consumers.

There are a lot of very good programs that have died a "freeware death".  A few months ago NDoc (the .NET documentation thing) died, and it was freeware. The interest level from the original developer dried up. Were there a financial incentive for him, it may well still be with us. This is a real shame when good software dies.

This is one of the reasons that I often prefer commercial software - there is a financial incentive, and I'm more likely to get support on an ongoing basis. This offers me some degree of protection when I come to rely on a piece of software. It's VERY expensive to switch from some software. (Been there - been burned very badly - not happy with it.)

As for competition, it's a very good thing to have, but it also can kill certain market segments inadvertently. There are many spaces that are essentially dead because of neighboring markets being over-saturated. We'll never see any more innovation in those sectors unless the already dominant players make a move, e.g. Apple with iTunes becoming a store front instead of an MP3 player (whoopie!).

Personally, I'm hoping that the CLS (CLR & .NET) really have an impact in the Linux world because at the moment, the CLS is the only really good cross-platform spec. Java is almost laughable at this point. C? Too hard. C++? Too many issues. Delphi/Kylix? Abandoned. RealBasic? Interesting, but doesn't have the community that .NET or other languages have. The CLRs aren't quite there yet (i.e. Mono), but they're getting there. When it finally arrives, we'll be able to create true cross platform solutions that don't suck the lifeblood out of the developer trying to support more than 1 OS.

In the immortal words of Jello Biafra and the Dead Kennedys, "Give me convenience or give me death!" (I'm such a sell-out.)

[zridling]

Jello Biafra and the Dead Kennedys nail it. Look no further than AutoHotkey and how it irons out a lot of Windows annoyances and/or weaknesses. I'm impressed with people who have the sheer focus to stay with a single product for years, even well past a decade. For example, Don Lessau's XYplorer was a freeware program called TrackerV3 — for almost eight years before going shareware! Eight years! Me, on the other hand, I can't even motivate myself to finish painting the living room after almost two years.

Beyond those who use software that's 'complete crap,' I'm more surprised at those who build it. Don't they know their software doesn't even compare to others in the same category? And pricing is another topic, but I experienced the same with data analysis software throughout the 90s. It was sold to universities and student discounts were still prohibitive — $500 for the base, $100-$300 for additional modules, or just $2295 for the whole thing. I'd never be able to afford that, not even today! What was once great software like WS_FTP Pro priced itself out of the market. That's why I have "affordability" as one criterion on my Great Software List site. Who cares if your program is fantastic if no one can afford it? Let's hope that Microsoft fully opens .NET. I got a feeling they'll hold something back; it's their nature.

[zridling]

I meant to add that although there are some hotkey apps for Linux, something like AutoHotkey for that platform would be a lure for Windows users. I wrote to Chris Mallett asking him to what it would take ($$$) to build a version for Linux that just included hotkeys and hotstrings (awaiting response). Since time is money, maybe if enough Linux users parted with a dollar or two, it might be worth his while. Otherwise, it's a Hall of Fame Windows app.

[Armando]

I meant to add that although there are some hotkey apps for Linux, something like AutoHotkey for that platform would be a lure for Windows users. I wrote to Chris Mallett asking him to what it would take ($$$) to build a version for Linux that just included hotkeys and hotstrings (awaiting response). Since time is money, maybe if enough Linux users parted with a dollar or two, it might be worth his while. Otherwise, it's a Hall of Fame Windows app.

-zridling (May 03, 2007, 01:51 PM)

I know it's slightly off subject, and maybe you don't like the whole "Wine" idea, but.... have you tried Wine? AHT AHK [oups] seemed to work on Ubuntu Feisty Fawn with Wine-- haven't tried it fully though.

What I like about Wine is that, instead of porting an app., developers can just benefit from that "compatibility opportunity". I think that, in some cases, Wine should be the way to go.

[zridling]

Yea, WINE is fantastic, and for me, has worked even better than the cautious recommendation (about running all Win apps) everyone seems to give about it. However, the consensus is that you can get AutoHotkey hotkeys to work, but not hotstrings. Here's what Chris Mallett shared with me on a AutoHotkey Linux port:

[Chris Mallett]: I've copied Jonny on this because he's expressed the most interest in porting to Linux.  In this case, "porting" may be a misnomer because most of the code probably cannot be ported. This is because it's too heavily dependent on the WinAPI. Instead, the concepts and design could be used to create a program with similar features in Linux. However, I have no experience with Linux shells (front ends), so it would be a huge endeavor for me. But it might happen someday, especially since like you, I don't see myself being a Windows user forever.
 

[Armando]

[...] the consensus is that you can get AutoHotkey hotkeys to work, but not hotstrings. Here's what Chris Mallett shared with me on a AutoHotkey Linux port:

[Chris Mallett]: I've copied Jonny on this because he's expressed the most interest in porting to Linux.  In this case, "porting" may be a misnomer because most of the code probably cannot be ported. This is because it's too heavily dependent on the WinAPI. Instead, the concepts and design could be used to create a program with similar features in Linux. However, I have no experience with Linux shells (front ends), so it would be a huge endeavor for me. But it might happen someday, especially since like you, I don't see myself being a Windows user forever.
 

-zridling (May 03, 2007, 11:04 PM)

Thanks for sharing that, Zaine. 

[gjehle]

[zridling]

Here's news that the Linux encroachment (along with ODF) continues to gain momentum. Zimbra to offer Ubuntu Linux support:

Spurred by customer demand, Zimbra has completed a version of its open-source calendar and e-mail server software for Canonical's Ubuntu version of Linux.

Having good calendar software is a lure to quite a few folks, or conversely, not having it is a turnoff.

[Armando]

Here's news that the Linux encroachment (along with ODF) continues to gain momentum. Zimbra to offer Ubuntu Linux support:

Spurred by customer demand, Zimbra has completed a version of its open-source calendar and e-mail server software for Canonical's Ubuntu version of Linux.

Having good calendar software is a lure to quite a few folks, or conversely, not having it is a turnoff.

-zridling (May 05, 2007, 07:40 AM)

Very interresting indeed. The ability to "talk to outlook" and to understand "outlook talk" : priceless... Ok, maybe not priceless, but at least... luring.