We spoke with Michael Phipps during the holidays and asked to give us his view from the top, as he heads the Haiku project from his home in Rochester, N.Y. Michael is not big on words, as you will see from his concise answers to our questions about the Haiku-OS.org server, project status, the future, and more.

What's up with the server?
I got a call from Deej last night. The server core dumps on /bin/mount. So we are looking at other possibilities for hosting. The folks in .eu complained frequently that they couldn't connect to the website when it was working.

Can we expect to see haiku-os.org back online relatively soon?
Not sure what relatively soon means. We have to try to get the disks recovered and get hosting. I am hoping within a couple of weeks.

In terms of Haiku project progress, how much has changed since the summer, and has the outlook for the future improved?
The kernel has seen a lot of progress. So has the app_server. Incidentally, so has screensaver. Many of the other kits are getting fixes and such but not massive changes. The kernel is coming along - for example - we can run gcc (getting wrong results, but it runs).

With regards to the Kernel, what sort of changes have taken place under the hood?
Not so much changes as fleshing out.

At the last BeGeistert (013 in August), there were several demonstrations showing off implementations of fork(), exec(), waitpid() and so forth. Are these the kinds of improvements to the Kernel that you meant?
That and more subtle stuff. Like axeld is working on the file cache; it isn't exciting milestone stuff, but it needs to be .

Now that some time has passed for the project, and the team leaders have had quite a bit of time to organize their teams, how much organizational work has there been this year, if you could sum it up from a birds-eye view?
Our teams are pretty small and we don't do a whole lot to organize - when there are only a few people, you don't need to.

What behind-the-scenes actions have taken place to keep the project momentum going?
Momentum in oss comes from progress. That's where it is coming from.

Earlier this year you spoke about difficulties in finalizing a bank account for Haiku, the long overdue paperwork that was delivered, and talked about seeing a lawyer. What came of that?
90% done. The bank account exists. I have the legal advice I need. We got our end of the year books done; I need to submit that to NY state so that we can get the final paperwork done.

When do you expect people will be able to officially donate to the Haiku project?
They already can! We posted that on our website a while back.

Do you foresee there being a time when developers in our community can earn a living, working completely for Haiku?
Honestly, probably not. I don't think that people will donate enough to make that happen. It's also sort of tough right now - we are between big milestones.

How close are we to the next milestone?
I don't do estimates. :-)

For that matter, what is the next milestone?
I can give you an idea of what I would consider the next milestone. I would consider the next big step to be booting the kernel, app_server and net_server off of the HD using PCI/IDE and caching.

Into a graphical shell?
Maybe.

Do you have any news regarding the virtual memory system?
Nope. My personal life prevents me from doing any significant amount of coding.

Just over 366 days ago, you stated, "My guess, though, is that R1 will be "released", meaning that there is a burnable .iso on our website as a beta. I suspect we will be in serious bug fixing mode.", in an interview on OSNews... Now we know you were guessing, but just how far off the mark were you?
Well let's look at what it would take to be there. Basically, it would require more kernel work and more app_server work. We "lost" a major dev on the app_server. And axeld is only so clonable before you get DNA breakdown.

Ok, why are your answers so short tonite?
Nothing I can think of. The issues are more simple than they used to be. When I had 15 kits to talk about, I could talk a lot more. Now, really, there is a lot less to talk about.

How do you feel about the project these days?
So much stuff is at that mythical 95% state - game kit, input kit, media kit (except for encoding), print kit, device kit, etc. I feel majorly neglecting. I don't spend ANYWHERE near as much time on it as I would like. Working to pay the bills is taking a lot more of my time than it use to.

I don't get to code. I only can read my email on weekends. Pretty much, I can possibly only dedicate Saturday to Haiku. But these past months, I have been doing real life work on Saturday, too. That's a lot of the reason that there haven't been newsletters. I still have tons of stuff that I want to do and a strong desire to do it. But I just don't have time anymore.

How do you feel, emotionally?
Frustrated, weary, sad. Not because of anything anyone in the community has done, but because of my circumstances.

Mystical reality time... Picture a man, walking up to your door and handing you a plain white envelope that holds the key to make your circumstances better. What's in the envelope, and how can we, as a community, help make more of it happen?
Nuclear warhead?

No, really.
Seriously - enough money that I could hire myself and the dozen or so developers that we need for the next year or so.

At this point, looking at the murky future, are you like Johnny Nash, "I can see clearly now", or are you like CCR, "I see, a bad moon a-rising."?
If anything, this project has NEVER been predictable. When I think that everything is going well, bad things happen. When I am depressed, good things happen. I guess I should just stay depressed. :-)\

A few months ago, the SkyOS project announced they were going to use OpenBFS. What was your reaction to that, and do you foresee, or have you heard, of other projects using similar technology from Haiku?
Cool. Yes. :-)

Seriously - I am very cool with what SkyOS did/is doing. They have found a couple bugs for us. The *whole point* of an MIT license is to allow stuff like this. Not even allow - encourage. As far as other projects - Cosmoe comes to mind. A number of commercial ventures (who want to remain quiet) have also approached me.

When are the newsletters coming back? If not by you, perhaps there could be a call-to-arms for some talented individuals in the community who could help?
I want to shoot for 1/1/05. We have some articles (waiting on my editorials). Could definately use more.

What am I missing? Is it -really- that far between milestones? Surely media kit.. networking.
Networking is tough for milestones. It works or it doesn't. The media kit hasn't really moved a whole lot; some new codec work was done, but the major contribs had crazy stuff going on in r/l.

So basically... a LOT of our developers are tied up in RL?
Umm - yeah. To a pretty good degree. That's why putting a few people on salary would make a majorly huge difference.
Don't be too negative, though. This happens every so often. In other projects, you don't notice it because there are more developers.

For example, mikew from the translation kit is helping out with the interface kit stuff. So there is an example of a finished kit donating devs to an unfinished kit. Also, the splitting of the community. You have some very good devs off on other projects (java, bezilla, zeta, etc).

How many active developers (devs who contribute at least once every 2 weeks) does Haiku have at this point?
Even every 2 weeks is tough. Some people submit rarely but in big chunks. (darkwyrm, for example) The number of devs hasn't really changed. No one has quit or anything. It's just that real life has interferred with a number of them. One dev is a teacher; he only really codes on school breaks, but then he codes like 12 hours a day...

Can you give us an idea of how much financial aid has been donated to the project, and where the money has been allocated?
We will publish this when the final report is together. I can give you a rough idea from memory - around $5000 came in. Most of that went to Waltercon (and was given for that reason). We have several hundred dollars in the bank.

Let's fast-forward a few years... I imagine by this time Haiku R1 is out, and making some people turn heads. Have you considered the marketing responsibilities, and how they will be handled?
Yes. :-)

Honestly, I don't think that R1 is the release that we want to *really* market. Because, to some degree, many of the things that the anti-haiku folks say are true. R5/R1 is 5 years dated. There are many things that need updating/fixing. R1 is not the release that I want people who have never seen BeOS to try out.

I really want R1 to be something that BeOS/zeta users use, enjoy, debug, test and bask in. I want R2 (or maybe R3) to be something that you show to people to make their mouths hang open. We have always said that R1 is a stepping stone.

What is your most pressing responsiblity in real life now?
Feeding my family. Well, that is a euphemism - making enough money to pay the bills, etc.

How are you feeding your family?
I am selling durable medical equipment. (wheelchairs, walkers, etc)

You're not one for words, I know... Can you tell me what pitfalls you've learnt over the years, and how you would avoid them if given the chance to do it all over again?
I have honestly screwed up more than anyone has any right to. And I am so aware that I can tell you probably every mistake from the beginning. Probably, over all, I have been too optimistic. From the very beginning, I made the mistake of accepting what people said they could/would do as gospel. Trying to grow an organization from scratch, I put people in positions that they weren't as ready for as they thought that they were.

We have really moved away from that to more of a "earn your position" kind of methodology. You want commit privs? Send us some patches. You want to be a team lead? Be a good team member first. That's how most OSS works. But, then again, most OSS teams don't need 10-15 leaders to start with like we did. No one has ever (AFAIK) done what we are doing.

At the beginning, when we had 100+ developers (that is, people who signed up to help), I thought that we would be done in a year. Which might have been the case, had all 100 devs been effective. But instead, more like 10 have been effective. That's why we are here 3.5 years later - lack of dev free time. Which comes about from a) lack of devs and b) RL obligations. Both of those can be fixed in only one way easily - money.

You know the saying "It's lonely at the top."... Is this especially true of an open source project, given also that there is a heavy responsiblity to re-create a beloved OS with an almost cult-like following?
Not at all. Because we are OSS, we work in a big glass house with no curtains. Because anyone can look at our source and see how things are going, there isn't the lonliness that comes from having secrets. I actually feel like I am part of that cult like following.

What kind of work has there been on USB support? What is expected?
I haven't heard anything lately on this. Need to check in with niels soon. I expect that we will have a USB bus manager and at least a few drivers. (hd, print, mass storage)

Will there be a WalterCon in 2005? If so, what is planned, and where?
Yes. My only real plan is that I would like to do it in Vegas. Something a little bigger.

In your opinion, what is Haiku's biggest strength over Linux?
I think that is like saying "what is an orange's biggest strength over chicken".

Seriously - Haiku is not competition for Linux. We are shooting for the areas that Linux is very poor at. I have been using a BSD as my desktop on my portable machine. I have come to understand, firsthand, why people say that Linux and BSD are not ready for the desktop.

I am a unix guru (10+ years) and I struggle hard every time I have to change something to figure out which text config file I have to tweak. That is where BeOS really stands out. Easy to set up and run.

Looking over 2004, what stands out most in your mind, both positively and negatively?
Positively - WalterCon, 501(c)3 status, booting off of a hard drive, input server finished. oh - and being laid off.

Negatively - transitioning from disposible income to broke, transitioning from being able to work on Haiku when I wanted to -> working all day.

What's close to being done soon?
Well... there's so little that isn't done that there is nothing coming.

Literally? The interface_kit, kernel, app_server, media_kit, networking_kit .. everything else is DONE!?
Mostly. I mean - we have bug fixes and stuff that we need to do. But most of the rest is done, AFAIK. Input kit works. Game kit and the new media kit together work. Translation works. Mail kit, Tracker. The situation is really a good one to be in - we are really down to the last few (hardest) pieces.

That about wraps it up, Michael... Anything you wish to say to the community at large?
Help.

Seriously. Now is the time. If you believe in Haiku and want to see a future for it, now is the time. Code if you can. If you can't, give as much as you can. We will put that money to work accelerating the development of Haiku. There is a window of time in which Haiku will have the chance to make a real difference in the world.

That time is slipping through our fingers. Slowly, torturously slowly, the other desktop os's are getting better.
At some point, there will be no compelling reason to switch away from whatever OS to Haiku. Now is the time - the pivotal point. If we don't get R1 finished and get on to R2, there won't be a point in bothering. If you believe in Haiku and want to see it in 2005, do something about it now.

Thank you Michael. Haiku News welcomes your concise, honest answers to our tough questions, and we hope our time spent will make a dent in informing the public at large as to the current and future status of the project. My advice to you, if you will take it, is that all good things come to those who wait, and work hard. You have certainkly done both. Thank you.

I enjoyed your questions, as I always do. :-D If anyone has any further questions, they can feel free to write me (mphipps1@rochester.rr.com).