Slashdot Mirror
Interview with Miguel de Icaza
GonzoJohn writes: "Linux Orbit editor-in-chief John Gowin contacted Helix Code to catch up on where their products and projects were heading in the New Year. Miguel de Icaza, GNOME evangelist and developer and Helix Code co-founder was kind enough to answer our ten question interview. Included in the interview is some new information on the Red Carpet Project, the next iteration of the Helix Update utility." Miguel also speaks here about the not-yet-feature-frozen Evolution (what happens when Evolution is declared "finished"? ;) ). Is anyone out there using Evolution in their own workplace?
I'm pretty sure emacs is done. God forbid someone decides it isn't and adds anything more.
Here (for the lazy).
Yah -- but doing that can involve a significant amount of extra work, depending on how the program's structured.
If the author uses a plugin architecture, then it's not an issue at all -- optional functionality can be compiled into standalone libraries, and those included in subpackages. Easy, done. However, if the options change the single, primary binary, then making subpackages to handle the problem is all but impossible -- even if you have them overwrite (or flat-out provide) the main executable, you can't provide a subpackage for each possible configuration of even a moderately complex package.
For an example of this problem, look at Red Hat's vim package. It includes 'vim-common', 'vim-minimal' and 'vim-complete'. 'vim-common' and one of 'vim-minimal' and 'vim-complete' may be installed. However folks can't pick and choose as they could with a source package ("I want the python scripting, but no TCL, and..."). Frankly, there's no easy solution that doesn't involve asking program authors to do way too much. After all, having a 'vim-python-only' and a 'vim-python-and-tcl' is hardly a workable solution.
Thus, while binary packages are a simple solution, folks who really care about their system's configuration are best off with source -- simple as that. It's hardly fair to blame the program authors, though.
You are truly confused about what free means, and what the Gnome project's goals are.
Why don't you just install Windows and Outlook? You'd be happier.
PINE is under a proprietary license by the University of Washington. You are not allowed to distribute modified versions. Please do not lump PINE and EMACS in the same category.
--
I noticed
--
I noticed
It's getting about time to leave everywhere
Hah. The thing that'll cook your noodle:
As a user, I'm free to use either Outlook Express OR Evolution.
Stating on Slashdot that I like cheese since 1997.
Why? Why would you want to store your OS let alone your data files on remote servers? What benefit do you get? I have SDSL here and I'd still kill someone before saving anything over 150kb on a remote server on a regular basis. Time! Time! Don't waste your clock cycles...
That's what I immediately thought, too. But it all depends on how it's implemented. You can simply have a button in the clip arts folder, labeled "More". This could take you to sites with downloadable clip arts, some of which have to be payed for. And you could simply do this from within the Gimp dialog. This scheme could be useful in a number of contexts. Depends on intent of the implementers.
"When I first heard Daydream Nation it quite frankly scared the living shit out of me." -- Matthew Stearns
This kind of reasoning is simply dishonest. The fact is that the only reason to use pointer arrays in C++ is for performance. If you're satisfied with java=like performance, you can use vector and not worry about "array out of bounds" errors.
Python "answers" these concerns by providing container classes (like C++), and providing reference counting.
C++ has container classes which are managed automatically, so let's move on to reference counts. Reference counts are fairly easy to implement in C++. They do not free you of all memory management because it uses reference counts, and not garbage collection. This is a convenience for sure, but when one runs into hard memory management issues (such as writing a graph class for example), the reference counts don't make things that much easier. In fact C++ programmers usually use reference counts as a performance enhancement technique (since it saves redundant copying)
Python may well supplant perl one day , but it will not replace C++ because it is not and never will be as fast as C++. Its design goals are completely orthogonal to those of C++. It's certainly a great language, but you're kidding yourself if you think its performance is acceptable for speed critical applications.
Still, I disagree with the following comment the other guy made: The fact remains that Linux and Gnome are based on a language and runtime that is processor and architecture dependent, provides no support for runtime safety and fault isolation, and provides no dynamic type information.
CORBA has built in exceptions, C is an ISO standard and GNOME sits on top of various portability layers (such as glib). I'm pretty sure GTK+ has some kind of dynamic type information, CORBA certainly does.
A high learning curve doesn't mean that the systems are "hard" to extend or modify, it means that it's hard to learn to extend and modify. I know this sounds pedantic, but it's actually an important difference. I'd argue that a reasonably proficient C++ programmer shouldn't have much trouble working with KDE, for example. I'd agree that it's quite difficult to become a reasonably proficient C++ programmer though.
Aren't MS using some big virtual machine for their common runtime ? I'm not that familiar with it, pardon my ignorance. I'm suspicious of anything that sits on top of a VM. (I'm concerned about performance, mostly ) I'm also suspicious of GC. I can see them pulling off a good imitation of Java, with slow performance as part of the deal.
I didn't really get what you were trying to say about GNOME. Are you trying to say that CORBA doesn't have exceptions, or an object model or type information anything like that ? IIRC, GNOME has all the things you claimed it didn't have (though I don't program in GNOME that much -- I prefer KDE). QT certainly has its own rtti system, its own object model, and a CORBA-like system (DCOP) which acts as a cross-language object system.
GC, virtual machines, etc. Is it going to need a Pentium 4 to even run ?
Bonobo doesn't even come close. It doesn't provide a common runtime, fault isolation, or garbage collection.
CORBA provides reference counts and exceptions. I believe it also provides runtime type information.
Bonobo would basically just give you the same functionality COM/DCOM give you, with all the problems that that entails
Such as ?
Can't get to page, it says the page doesn't exist.
--
Computers are useless: they can only give you answers. -- Pablo Picasso
What I meant was that Evolution is free software; i.e. you get the source code, can make changes, redistribute your changes, etc. Outlook Express is available binary only (free as in beer), and only runs on Windows (maybe Mac too?). If you want to run Outlook Express (or IE), Microsoft has already over-charged you for Windows, and possibly Office, so you really are paying for it. You don't have to pay Helix Code a cent to use Evolution, or Helix Code Gnome or anything else for that matter. I believe all their software is GPLed.
Outlook Express is not free in the sense Miguel is talking about (free as in speech). The cost of development was absorbed from the profits from Windows & Office, same as Internet Explorer.
Linux Orbit: Since Comdex, I've noticed a bit of a slow down on the number of updated packages on the Helix Code Update servers, is there a reason for this?
Miguel de Icaza: Yes, there is a reason. As we grow to support more operating system and more distributions, we have been working in parallel on a new system that will allow us to automate a lot of the work that we did in the past manually and to provide better quality assurance and support for our users. This project (called blue sky) has taken most of our resources recently, and the last batch of updates has been completely done with it, so we have been basically migrating from our old setup to a new, more extensible setup. For example, some users want to have a stable system and are just interested in getting bug fixes, while other users are interested in getting the latest and craziest inventions from the GNOME community. So part of our work is to establish various "levels" that can address the needs of those groups.
----
Celebrate the finer things in life
I just don't get the remote storage thing. Why would I bother storing files in a remote 25 meg space, when it's cheap to buy a hard drive with 25 gigs?
Off-site remote backup is important because your data can be accessible anywhere on the planet and more secure than it being on a secondary drive (as i know personally).
SEO Copywriter. Just Say ON
Among other things
SEO Copywriter. Just Say ON
Actually, Helix will use Nautilus. The reason is that Helix is a distribution of the Gnome environment.
I agree. And as you stated, Helixcode is under the umbrella of Gnome, so Gnome + Nautilus would remain correct.
Gmc is going out the window
GMC core dumped on me numerous times, so i won't be missing it. Good info though =^)
Nautilus isn't just some add-on file manager for the Gnome project. It will be the file manager for the Gnome project.
Good news indeed. But i've used Nautilus from the beginning, and it's far more than a file manager, unlike Red Carpet which seems like a package manager on steroids.
SEO Copywriter. Just Say ON
If open source efforts want to compete with
I'll agree Java is designed to make it difficult to code illegal operations. C++, on the other hand, is a language considered unsafe because it allows unchecked array accesses among other flaws. But i believe Python answers those concerns, and i expect Python will completely usurp Perl and C++ within the next 10 years. Already, many C++ programmers use the Python language for prototyping the full specifications of a product because it's syntax is so clean and readable.
The fact remains that Linux and Gnome are based on a language and runtime that is processor and architecture dependent, provides no support for runtime safety and fault isolation, and provides no dynamic type information.
Bullshit. It sounds like your an advocate of Microsoft Internet Information Server 4.0. Open Source supports everything you claim it doesn't. Except your precious Visual Basic.
It's depressing to me to see how much projects like Mozilla, KDE, and Gnome have followed in the Microsoft footsteps and are repeating the same mistakes.
Again, subtle M$ bullshit propaganda. Mozilla 6 couldn't have been constructed any differently than IE, and Gnome has always been radically different than anything M$ has ever conceived. Gnome should be considered a raging success, especially considering it's young age.
Happy New Year q000921
Make it your new years resolution to leave the dark side!
SEO Copywriter. Just Say ON
SEO Copywriter. Just Say ON
- make money from popular software like Windows and Office
- combat pirating
- widen it's reach to more computing devices
while Linux will uses this internet intergration to help users like you and i keep up with the warpspeed Open Source development going on around the world.I agree Open Source will use different technologies than MS, but the means is the same: software will be delivered as a service on demand to more devices than ever possible.
SEO Copywriter. Just Say ON
the "h" is missing from the start of http.
try this instead.
+++++
+++++
The harder you look the less you see. That's what we're up against.
On the other hand, to save bandwidth, or to simply work around having to download using a modem (oh so slow), I download Helix-Gnome at work, burn it to disk, and then install it elsewhere. At last count, there's probably about 12 machines using helix-gnome from a single download.
If installs of Helix Gnome are anything like installs of Linux it'll be near impossible to count the number of users.
Regardless, I think this misses the whole point. I don't use Linux because there are a lot of Linux users, I use it because it works. There are a lot more Windows users, but that hasn't made Windows a better platform has it?
Yea, miguel de icaza wants free software over usability, lets kick him off the evolution devel team !
...
<rant>
I mean, come on. This is a big part of open source, you get what the developers want. If miguel finds open source to be the most important thing to him, then thats what your going to get. You always get what the developer wants. OSS developers(not all, but ones like miguel) are doing this because they enjoy doing it, and because he wants to. Corporate companys do it because they are a slave to the dollar and do anything to make a buck.
So corporate america doesn't care, so what? While i like the ammount of applications and support linux is getting, and know that corporate america is where the "cash cow" of linux is, I really wouldn't mind if people started saying "Why do we need that, 99% of linux users dont want that, just corporate america, screw that and lets do something we want in our software."
I know this isn't the way to get linux "into every home/office/etc." So what? If your giving away your work, your doing it for a reason. His reason is because he beleives in OSS
</rant>
Well, I suppose there's still the faint possibility that somebody might find another bug in it, after all these years, but it's not getting any new features, and it hasn't changed in years.
.sig.
--Moss
This is a
Now there are two of them.
--Moss
This is a
Now there are two of them.
There are two _____.
Outlook express is *not* free. The opportunity cost is very high.
"Software like Outlook Express has been free for ages, so that's pretty much the norm (not the exception) for this kind of software."
Again, the word free shows its ugly head... For many, the source to freedom is the source.
The last time I did try it was rather crappy, esp. for slow IMAP mailbox windows.
The best one for Linux/x86 for the moment is the recently released mulberry.
Its a closed-source commercial product, but if you do IMAP, it really rocks your world. You can even adminster parts of the cyrus imap server with it !
Samba Information HQ
If it's so obvious, why does it only "most likely refer" to GPL-Compatible?
Secondly, plently of "free" code that's free enough to allow anyone to make contributions aren't "GPL-Compatible" - so it's pretty irrelevant what license is used. There are plently of routes that benefit from having available source..
-bugg
"For example, if you want to get more clipart in the GIMP, or get a pantone palette, you could buy those directly from the software you are using (on the color selector, in the font selector) "
Eww that doesnt sound good at all. I don't want an add popping up every time I try to select a font or color in the GIMP.
And just how are we to 'work to make this' anything if it's not free software? Don't like Miguel's focus? Grab the sources and start hacking. Can't write code? Get some cash together and pay a bunch of hackers to make it what you want.
If it wasn't GNU-compliant, you wouldn't be able to take the "let's make it x" attitude at all.
In contrast to your thought that having the software be free helps no one, I think it helps anyone--anyone who wants to take advantage of it.
Your post reads like more armchair quarterbacking than anything else.
I think you're missing the point of Evolution -- it's is GROUPWARE -- which Outlook Express is not. I personally cannot wait for Evolution to be finished, it looks like it will be one helluva groupware client. Now, what about a free, more-powerful alternative to the crappy Exchange 5.5 groupware server I'm running at work?...
According to Miguel, "you could say that Evolution is targeted to replace Outlook". Do you think they mean in terms of introducing new security holes, too? :-( It sure looks that way. Helix Code, like Microsoft, seems quite adept at crafting eye candy, but I'm not sure security is even an afterthought.
He did say the main site and there are plenty of sites that mirror the installation....so this could be right.
helix-update isn't doing it for me. I'm a Mandrake 7.2 user, and they aren't updating our RPM's well enough, and I'm using some stuff that is just simply old, like I was with that libgal3 that needed to be updated. For Mandrake, xmms 1.2.3 is still in the helix tree -- although 1.2.4 has been out and stable for quite a while. If helix is going to start packaging certain things, they must stay on top of it, and for ALL distributions.
Mike Roberto
- GAIM: MicroBerto
Berto
i maybe wrong, but i heard a while back that different front-ends could be made for it, also, you could run it under kde any even if its not a kde app :)
Do we just keep the same goal and play top-o-the-heap with other free mail clients? Do we now change our goal to "best mail client period? If the former is the case, then fine, but that will never meet the goal that Miguel clearly has of drawing Windows users to Linux. If it is tha latter, that's just bad project management. A project should have a clear goal and stick to it for the most part. Being the best mail client period for me certainly includes hackability and an extension framework, and other features that perhaps no current mail client has, and it doesn't mean parroting Outlook, which has it's own flaws and limitations.
Enough about me anyway, I was merely defending the original poster whose point was that when a user says "why can't I [insert feature] with this mail program" the answer shouldn't be "hey at least it's free software" it should be "switch to free software and you'll have that feature and more due the superior development force of the Open Source community."
All kings is mostly rapscallions. -Mark Twain, The Adventures of Huckleberry Finn
The best example of this is back from Feburary when there was an announcement on the KDE pages that said they were going to collaborate on the object sharing framework. A few months later, all mention of this vanished and KParts appeared as KDE's object model and Bonobo on Gnome's end. And guess what? They don't work together, and in fact the work that would be required to get them to work together would be tremendous.
I used up all my sick days, so I'm calling in dead.
2 beta pieces of beta software don't maketh a dot-net.
Miguel, Thanks for all the great work. I am a bit concerned that Evolution looks headed for an unstable release. It looks extremely promising and I am using 0.8 to handle my mail. But, it's wayyyy too unstable. It might not be a bad idea to set the counter back and label the release after 0.9 as 0.1.
get 0wned. irc.w30wnzj00.com
They did it themselves.I liked Eudora and I would've pay for it if it was available for Linux which it isn't.It was their stupidity.How long do you think Netscape would've last with their inferior to IE browser without their support for so many archs.? Sorry to say but Good riddance Eudora.
The best way to escape from a problem is to solve it. Alan Saporta
Yeah, it's called vi.
"We obviously need a new moderation category: (-1, Woo-fucking-hoo)" --Mr. AC
Anyone knows how is Red Carpet going to work? Is it going to be GPLed? It seems to me that it is going to be some great piece of software, that will solve all the problems of Linux packaging. (at least that's what him said!)
But I think it will be very bad if this great software works only for Gnome, as there a lot of people who use (and prefer) KDE.
I hope that the Helix team will do their best to keep the Red Carpet system as open as possible (as keeping their dependency database in a well documented xml format)
It would be great if the underlining tecnology were toolkit and desktop independent, and it only had a Gnome interface, but I don't know if its going to happen.
Anyway, does anyone has more information on it?
---- Email is reversed
ML had support for this same concept as far back as 1994 or so -- it too was free software, but it hasn't been maintained in a very, very long time. The web site is stull up though, for those who are interested: http://people.netscape.com/max/ml/
(Hmm... why aren't we all using StarOffice or ApplixWare?)
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
You claim that Outlook Express is free, but it's only true if you don't value your freedom. To me, and many of the people designing Evolution, freedom is more valuable than extra functionality.
And what about CVS access?
A journey of a thousand miles starts with a brutal anal raping at airport security
IMHO v8 is about 5 times as stable as 5 (skipped 6 - it was a dog)
and far more stable than lookout (tho spruce is still more stable)
Microsoft(tm) - a particular virulent virus that has infected most Pc's.
I could give a shit about my karma, I just was hoping someone could spread some light as to why in the world Helix Code Inc. would be using windows when their focus is to provide an alternative to it. I mean, if it isn't good enough for their marketing department, how is it ever gonna be good enough for mine?
apt-get install redhat please god - Me (take it easy, I love Debian)
Of course you know the reply to this is to 'get the code and do it yourself'. :)
creation science book
I understand the theory, but I'm not sure if your suggestion is on the mark or not.
"If one proves to be better, it will surpass the other. "
How can that be, if there is no agreement on standards? The fact that KDE can do something the GNOME can not is not evidence to a majority that KDE is 'better', so how can it 'surpass' GNOME?
MS Word is dominant now not so much because it 'surpassed' Word Perfect in terms of features. We had competition in that market - but that didn't spur them both to be better products. One got consumed by the other due to other forces.
It seems that the Linux community will adamantly opposed ANY efforts to get rid of Gnome or KDE or merge them precisely because many people WANT this perceived competition. How many improvements in KDE or Gnome have come about because one team looked at features in the other and said 'Hey cool, we'll add that?' - this isn't rhetorical - I've not followed the development of either that closely, so I don't know. Has it happened?
I thought simply having the source code itself, and taking input from other developers, was enough for a project to flourish. Now we need competing versions of the same thing to divide development time up?
creation science book
The "let's make this x" attitude could quite happily coincide with non-free software or non GNU-compliant packages by having the company or person that wrote the software accept user feedback as part of their development process.
Review and feedback is a tenet of the bazaar development model, but it's not exclusive to GNU-compliant packages. There has been many a closed-source developer that still listened to their users and made improvements that were requested.
Indeed, one could make the argument that someone developing commercial closed-source software has more of an impetus to do so, because if it doesn't do what people want, it won't sell, and if it doesn't sell, the developers don't eat.
In the poster's point about email clients, whether it was open or closed source, end users providing feedback may be all that's necessary to make it the best email client. To be certain, perhaps most of the users can't even read source code to begin with, and so could only possibly give feedback anyway.
creation science book
Good one! Alright! I've got quite a nice life, actually.
Not sure what qualifies that comment for 'wankerhood' - goodness! Anyone else who posts 'grab the code and make it do what you want yourself' is normally applauded as some sort of hero for having the 'guts' to think that everyone who uses a computer should be a kernel hacker. Don't know what changed in this case...
creation science book
Many people seem to think so, but it really wasn't... It started off as an excuse to test a new app framework we put together, just to see how quickly some type of basic interactive thing could be put together (regsitration, etc.). Didn't turn out too bad for a few hours work. :)
creation science book
Is there any software for Linux that is NOT in development?
As far as Agent goes it suffers from market saturation-"free as in choice", as I've heard others say about Linux. Xnews has taken many a user away from freeagent.
What idiot moderator gave this Score:1, Informative ?????? It's obvious that you didn't click on the link. And I would advice anyone NOT TO CLICK ON THAT LINK. This is worse than the goat stuff. The AC and the moderator that modded this up should both be put against a wall and SHOT.
Nobody believes the official spokesman, but everybody trusts an unidentified source. -- Ron Nesen
all software in existance is still in development. :)
thats one of the first things you're taught when you study software design.
so, in answer to your question....no
i wish i was but oh well
I don't think that Helix Code is bound to use any one of the hundreds of programs available at gnome.org. When I look at the the Gnome.org Get more software I see a lot of things that Helix does not distribute. Helix is a distribution of Gnome, as such, they are free to choose which file manager to ship. If Nautilus competes with Helix Code's own package management system, which is tied to Helix Code's business model, why wouldn't they leave it out, and include another file manager in their distribution of Gnome?
Looking through the gnome.org site raises a question for me. Where is the info on gnome 1.4? Maybe I didn't look hard enough, but all I found was a 6 month old reference to it shipping sometime in late 2000. I guess there is more current info on various mailing lists, but as I am not a programmer, I haven't subscribed to those. Can you point me to some links where Helix says that they will use Nautilus? I think I've seen various press releases by Eazel saying that Nautilus will be tied into Gnome core but I can't find that info on the Gnome site, or the Helix Code site.
Besides, there is still the way that Miguel beat around the bush responding to the question in the interview about Helix Code using Nautilus. He doesn't answer the direct question. "Will Nautilus become the default file manager in the next release of Helix GNOME?" All that question needs is a simple yes or no.
"Linux Orbit: Could you explain the relationship between Helix Code Inc and Eazel? Will Nautilus (the Eazel file manager) become the default file manager in the next release of Helix GNOME?
Miguel de Icaza: Both companies work on GNOME. Eazel works on the Nautilus file manager, and the GNOME VFS. Helix works on delivering services based on Evolution and GNOME in general (through our Desktop Improvements team and our Helix Technology Labs team)."
I guess Helix could ship with only the parts of Nautilus that don't duplicate what Helix is doing with Red Carpet, as all the code is GPL and Helix can distribute their own version of Nautilus. But all this is speculation on my part. I've just read what I found at the repective orgs and coms web sites. And what I read is that Eazel and Helix Code both have similiar business models. They want to sell package managment services; Helix Code through Red Carpet, Eazel through Nautilus.
Yes, I see that mc is "likely to be moved to extra apps." I don't see that Nautilus is on the list. But whether or not Nautilus will be the default file manager in gnome 1.4, and I am trusting that it will be, doesn't answer the question.
"Will Nautilus become the default file manager in the next release of Helix GNOME?"
Just the fact that this question was asked in the Linux Orbit interview shows that not everyone assumes that Helix will ship Nautilus. Because Nautilus has features that make Helix Code's Red Carpet redundant, and Helix Code needs people to use Red Carpet in order for their business plan to pan out. What user is going to use Red Carpet for package managment when the same functionality is built into their file manager?
So what's with the question about Helix shipping Nautilus as part of their default? If Helix used Nautilus, there would be no point in having Red Carpet. I like how Miguel beat around the bush when asked this question.
http://www.eazel.com/catalog/info
http://www.helixcode.com/apps/redcarpet.php3
I don't think that Helix will use Nautilus. See my post (Red Carpet vs Nautilus) further down for why.
You can always compile the sources yourself to your liking. or take the risk on installing someone elses binaries
"I keep looking in the want-ads under 'revolutionary' but there don't seem to be any listings.. "
And I think Solaris and HPUX users are more likely to pay for Helix Gnome sot hats why they made native packages for those OS's
"I keep looking in the want-ads under 'revolutionary' but there don't seem to be any listings.. "
Not to mention the "cutting off the air supply" actions against Eudora and FreeAgent, unnoticed with all of the hullabaloo around Netscape.
GiraffeSville, a place anyone can call home
Sheesh, this goatse thing is so old now it is getting pretty boring. I assume your are not children anymore, in which case you are quite entitled to look at naughty pictures whenever you want. I know your excited, and you want to tell all your buddies, but just calm down little web wizard.
GiraffeSville, a place anyone can call home
But there's no reason to have both desktop environments installed on the same computer really... I can see the point of having the kdelibs etc installed on a GNOME desktop so that KDE apps will work, and vice-versa, but it's pretty redundant and confusing to novices to have lots of different but similar choices for environment, word processor, web browser, calender application etc on their own computer.
I'd hazard to guess that the average user would benefit more if they were using a distro that installed only what the distro thought were the cream of the crop in each of the application areas.
Jeremy McNaughton
------ Live simply so that others may simply live.
Now, I know I'm getting a little bit offtopic, but I think any distribution's ease-of-use factor for non-geeks (ie. people that don't have the time to fiddle and experiment, they just want to use a computer) if it included only one desktop environment.
And for that matter, install only one (or maybe two, but no more) word processor, one calender application, one way to dial out to the internet. It would make things a lot less confusing... most people don't think application centric on their computers, they think task-centric. Of course, this kind of thing would only happen bump-free with a universal file format.
Jeremy McNaughton
------ Live simply so that others may simply live.
palm-utils doesn't use it. yet.
Netscape, as you cite it, is really no better than vim -- the communicator and navigator packages each include a full replacement for the binary, rather than extra plugins loaded at runtime. If there were more options than merely "with/without mail/news/etc", this variant of the subpackage solution would be unable to handle it.
I agree that having better-architected apps is a Good Thing, too, but the overhead involved is often significant. Practical issues are involved, too -- I'm not about to go back and rewrite my old apps to use a plugin architecture, though I frequently do so with my new ones. A better interm solution is needed for today's users when they're in need of greater configurability. Right now, that's compiling from source.
Ah, then. Well, we agree on The Right Way.
Only thing is, when I'm writing an app, I architect things that way if it's conveniant. Frequently it is -- using a plugin-based architecture, if it's well-thought-out, is an excellent way to avoid the need to hack in features crudely. However, at times when that's not expediant, I'm not going to spend an extra few days that could be spent actually Getting The Thing To Work on modularising it just for the sake of binary packages.
That's a common problem with binary packages. If you compile Evolution by hand, it'll detect the lack of palm-utils and not use it. Done, fine, everyone happy. That the binary packages are compiled with palm-utils support in is their problem (unless you think it's important 'nuff to break off subpackages to deal with the problem). In short, it's an RPM problem, not an evolution problem. Use the source and you'll be fine.
SharkMail was doing the virtual folder (they called them persistent search folders) thing back in 1997. I think they were the first email client to have this feature, but I may be wrong. FYI.
Let's try not to let fact interfere with our speculation here, OK?
Thanks, Miguel. I've started to read those docs, and some of my worries are disappearing.
With a bit of luck I'll also find that there are also (going to be) non-graphic front-ends to generate the intermediate XML to drive the back-ends, so that the remote, non-graphic or scripted installer is not left out in the cold.
Keep up the good work, and a Happy New Year to you!
"The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
You answer your own question in an oblique sort of way. Helix Code is focusing on providing an alternative to Windows. It seems quite obvoius, to me, that the best way to do this is to hire a bunch of competent programmers and start cranking out code. Now please tell me why, in God's name, would they want to waste precious time and money training the said "puke" to use an entire other operating system and mail client and God knows what else when they can just use those resources to improve their software instead? How is this not obvious to you? I've never really investigated the market for clerical workers before, but I'm almost positive that the benchmark for proficiency is being able to use Word, Excel, Windows, and Outlook. I for one sleep better at night knowing that Miguel de Icaza is harnessing his formidable coding skills to improve Gnome than wasting a week teaching Wendy the temp how to recompile her kernel and word process in StarOffice. Focus on the ends, not the means, man!
I think there is a world market for maybe five personal web logs.
The vision of the Helix Setup Tools is tightly integrated with our vision of component-based programming and interface/contract-based programming.
p -tools.html) and if you are interested in our approach to component programming and how we think these things should fit, look at http://primates.helixcode.com/~miguel/bongo-bong.h tml. And finally, if you are hooked into an object name space and service location you might be interested in the Moniker white paper (http://primates.helixcode.com/~miguel/monikers.ht ml).
The whole vision of GNOME is to enable a level of scripting that has never been available before. If you are interested in the technical details that address your concerns you can read the Helix Setup Tools white paper (http://primates.helixcode.com/~miguel/helix-setu
Best wishes,
Miguel.
I was using filter expressions in Elm in 1991. It provides a view just a normal mail folder, but that of mail that matches the filter. The only difference was that Elm's weren't persistent. It's not a terribly new concept, and it certainly isn't going to revolutionize e-mail.
--
I've finally had it: until slashdot gets article moderation, I am not coming back.
actually, Helix will use Nautilus..see my reply to your previous comment.
----
Celebrate the finer things in life
I'd just like to point out the obvious; that you can easily extract the source tarball from an SRPM. Use 'rpm2cpio foo.rpm | cpio -i'. rpm2cpio is included the rpm package, or there is a seperate version written in perl here.
I don't disagree that helix is ignoring slackware, but it is a fairly small hurdle for anybody willing to build from source.
He already made the VC pitch, and a VC firm listened. Linux Global Partners. They're really a great bunch of VCs, who are really with it.
--dave
Window managers and applications that are dynamically and effortlessly updated is a good thing, and believe its the future of home computing.
SEO Copywriter. Just Say ON
What are you advocating for? You want Evolution to be better than Outlook, why? You more users to use GNU/Linux from Windows, why?
Why would you want users to switch to another operating system if its just another propietary OS that restricts the user from the software by its EULA?
It is not just about being popular! There is the reason GNU/Linux is as nice as it is; there is a reason we have an entire community of contributors changing and sharing these changes across the internet or across the room---it all started long ago in this project called GNU. This kind of dynamic and helpful community only happens when the users have the freedom to do so.
That is what freedom is about.
Now I entirely sympathize with the fact that people often can't use free software because it isn't a viable alternative for them. But an alternative that isn't free is not alternative at all.
That is what Miguel was talking about. Freedom first.
"Red Carpet is a universal package manager. It has a pluggable architecture for doing package management and to handle dependencies in them."
It's like they say about standards, with so many to choose from...
Hate to say this since I'm a KDE guy myself... but I talked with the Helix guys at Comdex and they seem to be quite cool. Specifically, I mentioned the Slack problem to them and they kinda groaned, but Aaron came up to me and gave me the following URL:
http://primates.helixcode.com/~aaron/slack.html
If you want to go check it out, and if you REALLY want to try out that purty GUI installer, just follow his directions... and soon, you'll have Helix on Slackware. (I'm currently being tempted by the Dark Side - FreeBSD is awfully nice... ; )
Good question. Maybe she read the "It's in development" warning.
"Even though it has a non-zero version number, this is not a "stable" release. You will not be able to use it as your real mail client, calendar, or contact manager."
Or the README, which says something like "This program may delete all your email if you aren't careful"... or something.
"Can of worms? The can is open... the worms are everywhere."
I think you're trying to say "if the source wasn't freely redistributable.." - please don't mix the terms up; free and open source code predates the GNU project and the FSF by years.
Freely redistributable (in original and modified forms) source != GNU.
-bugg
What non-free software has taken off on Linux that had even one half-as-good--or even just promising--free competitor? This is a major
hurdle to overcome.
Games. Right now, closed source games are vastly mmore popular to Linux users than Open Source games.
VMWare, Win4Lin, StarOffice and Netscape 4.x are also popular closed source apps, vastly more so than their Open Source counterparts.
I think your vision of the Linux market is limited to yourself.
I guess `half as good' is fairly difficult to define. I think Plex86 currently falls in to the category of being half as good as vmware. I think Linux users will use VmWare over Plex86 for at least another two years.
There are many Open Source first person shooters which easily fall in to the category of `half as good as quake 3'. There are stacksa of OpenGL 3D games being produced by under Open Source licenses, most of which [same as their closed brethren] are crap, and a few of which are good.
But Quake 3 is, in my own observations, much more preferable to Crystal Space, or the Open Source Quake 1, or any of the other OS 3D FPSs. Its closest competition is Unreal Tournament, followed by Soldiuer of Furtune....get the picture?
Generally, Civilisation's genre isn't as popular due to the turn based nature of gameplay [most modern strategy games are real-time based]. Nevertheless, I'm quite sure the Open Source users that plays lots of Civilisation use Call to Power over Freeciv.
Although evolution works with --nodeps (for pilot), I'd rather just have it be more modular.
Mike Roberto
- GAIM: MicroBerto
Berto
What non-free software has taken off on Linux that had even one half-as-good--or even just promising--free competitor? This is a major hurdle to overcome.
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
VMWare had no free competition in sight when it became popular. I gather that Win4Lin is similar to VMWare, so the same applies. We will see whether this changes as Plex86 becomes usable, although head-starts carry some weight.
StarOffice, in my impression, is not that popular. I doubt it would have lasted long as proprietary software (and it may not last long even now that it's free). Most people in my experience use it grudgingly for MS compatibility. Again, no free program offered decent MS compatibility until recently (even now, many Excel and Word documents are not read properly by any free software I know of).
Netscape is an obvious case: there were no decent free graphical browsers until recently. Netscape had a long reign as the only reasonable option, which gives it lots of momentum. If Netscape 4 and mozilla were both new today, Netscape 4 wouldn't have a chance (on Linux).
Games are harder for me because I'm not much of a gamer, and because there are so many genres of game (not to mention that individual games periodically create their own genre). What proprietary games are you thinking of that have taken off on Linux? Shooters, and action games in general, had no free competition (Doom and Quake were eventually freed, but under unusual circumstances; I don't know how this has affected the popularity of newer proprietary shooters, do you?). The best case I can think of for your side is the Civilization genre. Freeciv qualifies as "half-as-good", and is quite popular, but I don't much about the popularity of the Loki Civ ports. You may score a point on this one, but I didn't doubt you'd find some example.
Here's the central argument: the free software community has two desktop/office projects that, while immature, clearly have basic functionality, strong developer support, and "long-term credibilty" (to use a term from the Halloween Document). Under these circumstances, I claim that no proprietary product that significantly overlaps with them will be successful on Linux.
The evaluation of an action as 'practical' . . . depends on what it is that one wishes to practice.
Judging from other GNOME software 0.9 isn't necesarily the latest release before 1.0., it might very well be followed by 0.10 and so on.
But in any case version numbers don't mean a shit, often get the feeling that people actually believe that these numbers actually symbolises some objective measurement of a project's current status.
You know what kills me about Helix? My LUG got an email from them a while back, take a look at the header. Thats right, they use WINDOWS. I realize she is just some marketing puke, but wouldn't you think a company formed around the idea of *nix software would atleast eat their own? If Evolutions is so great... Why arent they using it themselves?
6 0"
Received: by mercury.shreve.net (mbox rrowell)
(with Cubic Circle's cucipop (v1.31 1998/05/13) Mon Dec 11 16:44:07 2000)
X-From_: sblug-list-admin@sblug.org Mon Dec 11 15:42:39 2000
Return-Path: <sblug-list-admin@sblug.org>
Received: from server.sblug.org (IDENT:root@server.sblug.org [207.78.169.10])
by mercury.shreve.net (8.9.3/8.9.3) with ESMTP id PAA14391;
Mon, 11 Dec 2000 15:42:37 -0600
Received: from server.sblug.org (IDENT:mailman@localhost [127.0.0.1])
by server.sblug.org (8.9.3/8.8.7) with ESMTP id PAA02739;
Mon, 11 Dec 2000 15:42:03 -0600
Received: from trna.helixcode.com (IDENT:root@trna.helixcode.com [140.239.238.2])
by server.sblug.org (8.9.3/8.8.7) with ESMTP id PAA02725
for <sblug-list@sblug.org>; Mon, 11 Dec 2000 15:41:13 -0600
Received: from Default (michelle.helixcode.com [140.239.238.51])
by trna.helixcode.com (8.9.3/8.9.3) with SMTP id QAA03420
for <sblug-list@sblug.org>; Mon, 11 Dec 2000 16:41:00 -0500
From: "Michelle Scappace" <michelle@helixcode.com>
To: <sblug-list@sblug.org>
Message-ID: <NEBBLLHEALBBOAFKKNMFEEAKCCAA.michelle@helixcod e.com>
MIME-Version: 1.0
Content-Type: multipart/mixed;
boundary="----=_NextPart_000_005D_01C06391.6EBA88
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2910.0)
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Importance: Normal
X-MS-TNEF-Correlator: <NEBBLLHEALBBOAFKKNMFEEAKCCAA.michelle@helixcod e.com>
Subject: [Sblug-list] LUG
apt-get install redhat please god - Me (take it easy, I love Debian)
Here's a quote from the article that bothered me:
I am a big fan of free-speech software, but unfortunatly, to 99% of corporate America, free-speech is not an important feature for software. The goal of Helix Gnome and Evolution are to make Gnome a viable alternative to Windows. I understand that to mean that Evolution should strive to be absolutely better than Outlook (since it can't continually play 'catch-up'). 'Better than outlook' would be it's most important feature, since as I have previously mentioned, corporate culture doesn't care about 'free-speech'.Keeping
Bonobo doesn't even come close. It doesn't provide a common runtime, fault isolation, or garbage collection. Compilers don't generate type information for it. In fact, as far as I can tell, even if it were implemented fully and perfectly, Bonobo would basically just give you the same functionality COM/DCOM give you, with all the problems that that entails.
It's depressing to me to see how much projects like Mozilla, KDE, and Gnome have followed in the Microsoft footsteps and are repeating the same mistakes. To me, Microsoft is bad because they are using 20-30 year old technology for their GUIs and object systems. Cloning their implementations of outdated technology isn't going to propel Linux into the 21st century.
The fact is that there isn't a single piece of software that makes everybody happy. The problem with Windows is not that it's universally bad, but that it wants to be the one platform everybody uses. And I think it's not all that desirable that Helix/Gnome and KDE are trying to give us more of the same stuff. Using KDE2 is almost like using Windows now, down to the senseless editor bindings, and Gnome seems poised to follow suit.
The thing that worries me Miguel is that a lot of functionality is going into graphic apps directly, instead of going into utilities that are managed through graphic apps. Since graphic apps are almost entirely non-integratable using linguistic glue, we're in severe danger of losing touch with the key element that made Linux/BSD/Unix the powerhouse that it is, namely programmable integration through non-graphic scripting.
Your comments about Red Carpet brought this to mind vividly, and raised other spectres as well. We all hope that Red Carpet will become a great generic package manager, but alas it seems that the power user that is doing remote or scripted non-graphic installations is not going to be able to make use of your good work.
Frankly, graphic-only apps suck, or more technically, are not as powerful as graphic apps that interface to underlying non-graphic utilities. Why is Helix going down this non-optimal road towards Microsoft-style systems of low intrinsic power? Why not have your cake and eat it too by using graphics for interfacing only, not for implementing new functionality?
"The question of whether machines can think is no more interesting than [] whether submarines can swim" - Dijkstra
Well, I originally heard of query-based folders in the Paquiderm mailer (Jim Gettys showed this to us a long time ago). Paquiderm, and the help of Nyao Nguyen (who was a VM author). Jamie Zawinski later pushed for the current setup which is: work the way people expect using files-as-folders and add vfolders, and let people migrate.
Now, I did not claim it was a new invention, just that we hope that Evolution will help popularize this way of handling e-mail.
Miguel.
As I said on the interview (but it seems the url got lost somewhere), we have been working on a cross-platform set of tools to configure and customize your Linux system.
You can see Arturo's screenshots here:
http://primates.helixcode.com/~arturo/hst
Miguel.
An Intereing note, As much as they spout 'Free! Free! Free! Source Code!!', HelixCode's site does not have Any packages in either the SlackWare .tgz format (big deal, most slack users like to compile from source), or Source Packages other than in SRPM format...
I can't speak for Slack users, but I can tell you that you're dead wrong about source packages for Debian.
This directory is full of tarballs and diffs; using "apt-get source [Helix package of choice]" with the appropriate lines in your sources.list file will get you a source tree.
Jay (=
It's not about "copying windows." It's about using the good ideas from a variety of platforms and paradigms, supporting and implementing existing languages, and providing freedom. If you don't like Visual Basic (I'm not fond of it, myself), there is no reason why you have to use it. However, if you want to read an Excel spreadsheet in Gnumeric that uses VB scripting, the gnome basic support that's integrated with Gnumeric comes in handy. And if you're a windows programmer familiar with Visual Basic, this allows easy migration to the GNU/Linux/Gnome platform. Once there, hopefully you'll see the merits of other languages such as Perl and Python.
----
Celebrate the finer things in life
Miguel de Icaza: We have counted around 450,000 installations of the full Helix GNOME desktop from our main site. We have also distributed around ten thousand CDS. So we figure we have over half a million people using Helix Code GNOME now.
I realize that "total number of Helix GNOME downloads" and "users" is difficult to estimate, but does anyone else think that 500,000 is an overestimate?
450,000 installations & 10,000 CD != 500,000 users (No, I'm not arguing that the simple math is wrong).
Many of those installations are probably reinstallations. I've completely reinstalled Helix-Gnome onto this desktop right here 4-5 times.
I did the newbie http://go-gnome.com thing once, I downloaded & installed the RPM's manualy (After accidently deleting/overwriting something or trying to satisfy a mysterious dependancies for some some nifty-sounding-but-experimental package), and I completely reinstalled the entire thing from the source on Saturday. So that's 4-5 installations, yet I am One user. (And I did the same thing to my work Computer, but I imagine that counts as a second user in their stats).
Many of my Helix Gnome friends do this thing (But less often then Crazy-agressive-maybe-reinstalling-will-make-pilot link-and-esound-work! me).
"Can of worms? The can is open... the worms are everywhere."
If Miguel's goal is to make Evolution very "windowsy", is Miguel going to add in Gnome Basic VBA scripting abilities? It will be very interesting to see just how far Helix takes the whole "let's copy windows" thing.
Given, Slackware isn't the #1 distribution, or anything, but if they've gone through the effort to port to Solaris and HPUX and package for them, they ought to make source .tar.gz's available... After all,the Helix gnome is a lot prettier than the normal GNOME one... And Slackware users shouldn't be left in the dust here just because .rpm (and to a lesser extent .deb) is overrated...
get 0wned. irc.w30wnzj00.com
Virtual folders have been around for a while, among other places, in the Emacs VM mail reader. It would be nice if open source projects would acknowledge other open source projects.
(Gee, Miguel is beginning to sound like he's making a venture capital pitch.)
For many years i used pine and/or emacs as my mail reader. I tried kmail and it was not too bad for a while, but I don't run the K Desktop exclusivley (I run blackbox...roar) and K apps aometimes suffer negative effects from being run outside of the entire K environment. So it was back to pine
I tried Balsa for a while, it was pretty, but at the time, it was not threaded and died, alot. Once again back to pine.
A month ago or so, I decided to give Evolution a try. I must say, it is one hell of a good mail client , yes it crashes once in a while, but I just start it up again and there are no corruptions or anything. The mail filtering system works really well. The user interface is dead simple to set-up, and *heck* it's pretty.
I couldn't begin to compare it to MS Outlook (or Outlook Express), since I haven't used that mail client in many years.
But from a guy that has used alot of the new email clients kicking around, and has always reverted back to good ol' pine, Evolution is my mail client now.
Of course there is a soft spot in my heart for pine, it's still configured to read my Evolution mail box (easier for remote mail checking). And the uh *calendar* I dunno, it looks pretty, but I'm not a big calendar user...I prefer mass disorganization in that dept. *grin*
From a pure free software perspective, Evolution is designed to be the best mail and personal information manager free software product
I'm pretty sure that most users, and especially those coming from the Windows platform, couldn't care less. Software like Outlook Express has been free for ages, so that's pretty much the norm (not the exception) for this kind of software
Having used Evolution for a while, I'm really, really happy with this product - it's the first viable replacement for the POS Netscape mail client I've ever seen. However, I think that a focus on how GNU-compliant the software is doesn't help anyone: let's work to make this the best mail client available anywhere, period!
I know of a lot of Solaris users who wouldn't mind paying a sizable client license fee for a working GUI mail client equivalent to Outlook Express but without the enormous overhead of the Microsoft product (or even the Netscape client, for that matter...)