RealLife(TM)
To be honest, this article is something I have been dreading. It's one of those situations, where you hope and pray that it gets better before you take action, but it hasn't.
Michael has landed onto some tough times. The injury from some weeks ago has healed well enough. However, RealLife™ has prevented him from focusing on Haiku. It involves his personal life outside of Haiku. That's all we know and that's all we need to know. With everything that has happened and will continue to play out over the next weeks, Michael has decided to cease the contract. In the future, he does plan to contribute to Haiku. For now though, he needs to take a step back.
So, what does this mean to Haiku and its supporters? During the contract, both Michael and our supporters have been outstanding. A wealth of important commits have been made possible and have ranged from Sandy Bridge support, to integrating WPA/WPA2 support (albeit missing some features), to resolving a slew of critical bugs that required both a talented mind and ample time. The financial donations have set a new bar and will allow Haiku, Inc. to continue advancing Haiku. In the first four months of 2012, over $4,000 USD has been donated. That brings Haiku, Inc.'s current finances to over $25,000. Needless to say, funding development contracts will continue. Thank you for making this possible.
On the topic of money, it's worth mentioning how this contract is being handled. At the beginning of the contract, it was decided to issue payments at the middle and end of the contract, based on the work completed. Prior to Michael's injury, the first half of the six month contract has been paid. Even though the contract was ceased before completing the second half, he will receive payment for the portion of time worked -- e.g., from the second half up to where he took time off for the injury. Michael will be compensated for time completed -- no more, no less. We feel this is fair for everyone involved.
This is not a failure of Michael, nor our supporters, nor anyone else. This simply is something unfortunate that happened. We offer Michael our best wishes during tough times and welcome him back when he's ready. We're here for you buddy.
Comments
Re: RealLife(TM)
As a contributor and supporter of Haiku I wish Michael all the best in his life outside of the project. I hope this experience does not stop any other developers applying for contracts, as in my opinion these solve many of the larger issues with our target of R1 release.
Michael's work and dedication has never failed Haiku and I would like to say a big Thank You!
Re: RealLife(TM)
Time to get 5 left bugs killed now for Haiku Alpha 4
Let Michael rest for now...
See you soon Michael..
Re: RealLife(TM)
It's sad this has happened - these things happen, and I wish Michael all the best and look forward to seeing him back in action.
Compassion and friendship aside, I think some clarification is needed with regards to how the funds were distributed. To me, the work/amount spent ratio doesn't add up. Michael wrote (http://www.haiku-os.org/blog/mmlr/2011-10-27_my_first_month_contract_work):
"Well, I'm keeping pretty precise track of my working hours and in the end I am only going to bill for actual hours spent. So rest assured that everything is going to be fair. I'd really be uncomfortable wasting any of that precious donation money (back in the day I was a donor myself), so I'm trying really hard to ensure that the investment in this contract pays off."
I also want to make sure Haiku gets the most for its investment. Can these records be disclosed to the donors?
Work on the contract began on Sept. 27th, 2011. From here:
http://www.haiku-os.org/blog/mmlr/2012-02-05_contract_paused_due_health_...
The contract was paused the first week of January, and little to no work has been done since then. That's a total of 14 weeks of development work. Michael was to be paid 2300 CHF/month, so that should equal 8050 CHF. I've already been notified by a Board member at Haiku Inc. that Michael could be owed over $10,000. However, 8050 CHF = ~$8780 USD.
I think a proper exact explanation is owed to the donors as to how the funds are spent...
As a side note, it's unfortunate, but hardly any of the priorities were tackled in the contract. Maybe 5/100 users of Haiku have a 'sandy bridge chipset', while the other 95/100 have a wireless chipset and need to connect to an encrypted wireless network. I've been trying to support wireless networking since 2009 with the original wireless bounty at Haikuware - 3 years and thousands of dollars later there still isn't a complete solution. Overwhelmingly, most users that try Haiku complain they can't get connected with Haiku. What is an operating system today that can't connect to the net? It's this type of vision that fails the Haiku administration. Give the users what they want. I am at the point of abandoning Haiku because it just takes too long to get a useable release. Even when the tools and money are available, issues aren't prioritized and resources aren't used efficiently. Perhaps the contract should have been more demanding; i.e that these goals/priorities or x,y,z goals/priorities have to be fulfilled for payment.
I know if I had a contract and got sick, or had any situation that prevented me from completing the work, I wouldn't give up or miss a deadline. I would have sub-contracted the work by now to keep everyone happy.
In any case, I hope Haiku Inc. has a back-up plan/will develop one to use the rest of the funds resourcefully, efficiently, & openly. Hiring another contracter to mandatorily fulfill the priorities listed in Michael's contract as soon as possible would be my recommendation. It's my assumption a lot of donations were made on the premise that the priorities initially identified and listed would be tackled/resolved to put Haiku on the fast track to R1.
Re: RealLife(TM)
quote: The contract was paused the first week of January, and little to no work has been done since then. That's a total of 14 weeks of development work. Michael was to be paid 2300 CHF/month, so that should equal 8050 CHF. I've already been notified by a Board member at Haiku Inc. that Michael could be owed over $10,000. However, 8050 CHF = ~$8780 USD.
FIrst, the keyword phrase was "could be owed...". "Could" as in "it is possible, not definite".
It is possible that I may have been mistaken (not the first time and unfortunately won't be the last.) on how much work needs to be paid. It won't be known for certain until Haiku, Inc. receives an invoice. At that time, I'll ensure that some type of news/announcement/email is published.
Re: RealLife(TM)
Too bad the contract work had to be stopped prematurely. I'd loved to have seen what Michael could have achieved in the full 6 months. The first three were very promising. I can now surf the net from both my computers via WPA encrypted WLAN and Haiku is much more stable than half a year back! And having support for sandybridge gives me interesting options for my next hardware. :) Thanks Michael, I hope you can quickly resolve the issues you're experiencing.
Unfortunately, I don't think a superb developer with the deep insights of Michael can easily be replaced. As has been said many times before: most of the easy bugs are fixed, what's left are the hard and tedious ones.
That said, I believe a contract to finish package management doesn't need that deeply knowledgeable a dev. Maybe, until someone can be hired to finish the real hard stuff in the system's bowels, this specific project could be contracted (if there's a dev for it, of course). That would at least move us a big step forward to beta.
Regards,
Humdinger
Re: RealLife(TM)
that's a sad day. Being payed about 2.000 euro a month for your hobby, sounds a little like winning at lotto (and having very flexible working conditions, he could also moved to an tropical island and doing the coding from there).
Indeed there is a substitute needed. It's a little disappointing, that michael didnt inform the community sooner. Haiku lost a lot of time.
A substitute is needed, but I guess it's not easy to find.
Perhaps Alexander von Gluck (kallisti) ? Since he seems very active in last time.
Re: RealLife(TM)
perhaps it would be worth a try, to announce on major tech-sites that haiku is offering a full-time job. Perhaps people would apply and send their cv's and their demands. And the core-haiku devs could meet with the candiates on g+ making a hangout, and trying to find the best candidate.
(In case that there will be indeed people interested in working for haiku.)
Re: RealLife(TM)
Maybe it would be better to pay a developer for reaching a goal...
Means when one developer meets his goal he will get a fund/prize/payment...
More like the bounty system... and give them a list of targets.
1. wpa/wpa2 support
2. sandy bridge support
3. html 5 and css 3 support
4. installing Haiku gcc4 hybrid only works on clean install now... (initialise partition or delete partition first). was working before..
5. It would be great if Haiku is playing audio and video after installation... ogg and mp4/ogg video
Looks like the bounty system but vice versa...
Don't know...
It is very sad indeed...
Re: RealLife(TM)
The open-ended contract work has never really worked or produced any tangible/'headline' results for Haiku. i.e printing support for Haiku's end users, i.e Wifi for Haiku, i.e Bluetooth for Haiku, etc. Pay for work complete or contract fulfilled (incomplete package management, incomplete web browser, etc. can't really comment on Posix signals).
Re: RealLife(TM)
Should have been a reply to kvdman:
It is all in the way you measure. The amount of work for the examples you give probably is worth much more than you paid and it made the goal much easier to reach. IMO opinion both kinds are needed. The groundwork which makes the project achievable, and is very hard to estimate. A finishing project that does the implementation could then be created.
Re: RealLife(TM)
I suppose. Maybe I'm just frustrated because I thought that this push would finally bring Haiku closer to R1 after following the project for ten years; and there are delays and development focus shifts yet again. I do think certain identified priorities need to be addressed though.
I know that the projects completed through bounties were done for next to nothing. However, there are limited resources; goal oriented, priced contracts/bounties are a more efficient use of limited funds (IMO). There are always people that are out of work, work for less money than someone else, work 70h/week instead of 40h and can take on extra bounty work for cheap, etc. Cirpi gives the example that many talented developers in Romania work for 200-300 euros/month; as opposed to ten times as much.
How about Haiku Inc. tender mandatory goal oriented contracts to bidding developers so we can finally have what needs to get done in Haiku finished.
Re: RealLife(TM)
Sorry about Micheals problems of health or whatever. It is probably frustrating for him that he could not complete the contract aswell. I don't know how many average hours/day he was devoting to it but too many could affect health or personal relationships. I would tend to agree with a bounty system for all the initial preparation-estimation work in a first part project, then contract for the second part. How do programmers achieve goals in paid work company situation? I suppose they benefit by having close association with other programmers on the same project and one head programmer directing priorities and coordinating (not one myself though). A bit like the kernel/threads in a OS. If Haiku can run so smmothly why can't this be seen on the human level that is reverse engineering-designing it?
I suppose programmers in open source living continents apart can still have association and help but it might not be the same and as immediate as been all in the same physical location, different time zones, sometimes language, although English is nearly universal.
I still think open source is neccessary especially in Haiku case but wish it could be coordinated/communicated better between programmers.