OSNews hosts today a mini-interview with Frans van Nispen of Xentronix and the Sequel OS, a new BeOS-like OS which aims to continue where the BeOS technology left off as opposed to being a direct clone. Most people will know Frans from the AtariST demoscene and from his BeOS involvements. In this interview, we will query the status of Sequel (which btw, I am not part of anymore, so don’t email me about it please 🙂 and the status of Xentronix’s Refraction and SampleStudio PRO.1. How is Sequel progressing? What can you tell us about the advancements made so far to the OS?
Frans van Nispen: Sequel is coming along fine. Only two areas on the kernel need some serious work, though we have started work on other areas of the OS as well now.
– we have a kernel that provides about 90% of the desired functionality
– our libroot implementation is reasonably complete
– integration of libroot and kernel is good though some work remains
– presently sequel boots to a cli shell
– we are testing libroot and fixing problems as they arise
– design work on the various userland servers has started
2. How big is the team working on Sequel? How is work distributed?
Frans van Nispen: Our team concists of 10 people currently, devided in people with various skills on various areas. All members have more than sufficient skills
in the area they are working.
Work is distributed by inquiring what has to be done next and which topics are running. We try to split the work to the capabilities of our team as much as possible. The big lines are set by me, but we try to work out as much as possible as a team.
So far, we have members with outstanding skills in networking, low level kernel work, filesystems, graphic systems, media, common application development, graphical design and user interfaces and web development.
Any one who is realy willing to join can contact me at [email protected] though only people with a clear motivation and some level of skill are accepted as we want a team that works closely together, understand and
respect each other.
3. Please tell us a few things about the aims of Sequel in the technological
arena. What kind of kernel, filesystem, UI, toolkit are you going to use?
Frans van Nispen: We use a small modular kernel. Our initial filesystem will be BFS.
The GUI will be like the BeOS one: concistent. We will also follow
the Be design of having kits, though they will be extended in many
areas and new ones will be introduced. For example, we will add XML
support functions.
I would say that our target audience would be the enthusiastic developer
for quite a while, which fits nicley for any timeframe for the period
that our OS’s will mature over. The audience may then spread towards
users of existing hobyist platforms such as BeOS and Amiga users, and
hopefully at this point we may get third party commercial application
intrest.
As regards to software – I would say that because of the developers
skills and because we’re taking the best bits from other media-oreinteated
OS’s like MacOS, BeOS and Amiga, the OS should be pretty good at anything
to do with media, but will have strong UNIX-like underpinnings.
4. How close to BeOS, Sequel is regarding its design and goals?
Frans van Nispen: Sequel will follow a similar design to Be in terms of key architecture, ie a
small kernel with much functionality being provided by kernel level add-ons
and userland servers.
We believe Be created a great OS, though while they got some things right,
they also missed some functionality that’s useful and probably should have
been there. For instance we will support mmap() which BeOS never did.
The kernel module support that Be used was good and useful so we’ve added that
already, although we support sysctl() for getting details of modules loaded.
We follow the Be design of teams/threads, but removed the 192 thread limitation
BeOS had. This shows how we’re not afraid to use the best bits of BeOS, though
if we want to improve on them we will.
5. Please tell us about your other BeOS involvements: a final Refraction release and the status of SampleStudio PRO? (screenshot of Sample Studio, here)
Frans van Nispen: Refraction has gotten a great boost with the last beta 6.1. I would say it
now is about 90% finished. The main things to be done now are creating a
brush editor, fix some pending bugs and write a lot of documentation. This will
take a few more months as I am the sole developer on this project for months
now.
Our target release for SampleStudio PRO will be around the release of Zeta. We
expect to start the test rounds on Zeta in about two weeks from now.
The direct to disk editing, or virtual memory system, is finished now except
for a pending bug in cutting a piece.
The VST plugins now run like they were natively added. Lots of bugs that were
introduced in the beta 3.0 (which came with the source release) are fixed.
We are now running through a list of options that will be added for the
final release with our new group of beta tester who are all established users
of this kind of software. Any new effects or processing functions will be
added as VST plugin, so other BeOS/Zeta programs can also benefit from them.
I’m not a graphics artist, but for copy/paste thingies I usually feel the lack of “paste as new image” option, hope it will be included in the next release.
(but then again, maybe I managed to miss this option somehow.)
anyway, I wish a lot of succes to these guys!
i notice there’s a pepper icon in that sample-studio screenshot. (is that pepper for beos? – does it match up with pe-on-steroids’ syntax highlighting?)
Guy.
Yes, that is Pepper 3.5 for BeOS (old version, the last BeOS version of Pepper, current Pepper version is 4.0.7). It was only beta, and it never got finished, because of multithreaded-oriented deadlocks that was happening. That screenshot is from my machine.
What’s the plan on drivers? I see OpenBeOS and Zeta facing the same lack of drivers as the original BeOS had, where only BlueEyedOS will not have to worry about this (by using the Linux kernel and XFree86). A team of 10 programmers won’t be able to support a lot of hardware.
AFAIK, Sequel is a hobby up to now. Supporting only the most important/most used hardware will be the main focus.
Why not put support into yellowTab and their work to continue BeOS? Are we going to end up with two seperate and competing systems?
any chance to see a pepper screenshot from that beta? (im a syntax-highlighting fanatic…) (and like pe very much..)
I wouldn’t worry about these guys competing, obviously Zeta is about to start taking orders any week now, while Sequel is aimed at something completely different talking months away.
What can be noted is though lack of drivers is an issue, Zeta already has some support due to codebase inherited from BeOS R5.
Drivers are popping up all the time, and as the community is actually growing (it has been for last couple of months) both of normal users and devs, we’re likely to see a continous growth of all that is needed for success again for BeOS.
Unfortunately most seem to wait for OBOS progress, but that is still a couple of months away… a couple of kernel hackers would make a huge difference though, but they don’t grow on trees.
Kudos 2 Xentronix for great products…
since they brought up sequl I figured I could get away with this….
http://www.blueeyedos.com/news/index.html
apparently today they have just anounced on B.E. OS that tehy have released their first live demo cd for the publics perusal…..they are looking for people who can mirror it.
Sequel looks to be making some progress, but what it actually delivers is very hard to say from the information on their web site. What type of API etc. are they going to use (will I be able to compile a R5 source file???). The people working on it are properly talented, but what/when they deliver something might be some time in the future (based on Eugenia’s, it took Be 10 years and 50 engineers….).
Their decision to stick with some hardware is a good move. Instead to trying to support at lot of stuff, it’s easier to go with “if you want to have 3D HW acceleration use ATI Radeon” (or whatever).
In terms of popularity, OpenBeOS, B.E.OS, and/or Zeta will probably have a bigger chance. I have no idea what the quality of either of these BeOS efforts are, but starting again with new effort???
I have the deepest respect for the people working on Sequel and why they need a hobby, but I really can’t see why they won’t go with either OpenBeOS or B.E.OS.
I just read the news on http://www.blueeyedos.com, and it sounds great. Wonder if they got rid of the complete mess Linux delivers? No offence guys, I sill don’t like the name. What am I going to say to my mates when they ask me what OS I use??? I use Blues Eyed OS….. I use Blue….. Hmmm I hope I get used to it!!
Hygge…
> I have the deepest respect for the people working on
> Sequel and why they need a hobby, but I really can’t see
> why they won’t go with either OpenBeOS or B.E.OS.
As far as I recall, Frans was on the OpenBeOS Interface Kit team before but left due to various issues with either the management or some of the dev team members (not sure which), so that safely discounts that option.
Finally! I have been wait for this.. Umm, looks like we can use BitTorrent to help the bandwidths, mirrors and etc. I even will leave it runs for someone to leech my bandwidths on the regular cable for a month or so.
Agree re B.E.OS…. I don’t like Linux/XFree combination but the philosophy behind what they’re doing certainly seems a lot better than most conventional Linux distros.
Build a homogenous environment and the apps to run in it.
I like that.
on the B.E.OS website he still credited as the interface kit member ( http://www.blueeyedos.com/about/organization.html ). But this may be because of his past contributions to the projects, because afaik he is no longer with B.E.OS (afaik being the key word here).
Anybody who’s background includes the Atari scene has to have his head screwed on right =).. I’ll wait for anticipation of playing with this baby (but PLEASE dangit, hardware 3D support is a MUST:))
Everyone and their mums calls SQL (Sequential Querying Language) sequel. Whether its MS or Pervasive or whatever.
Now that Windows Longhorn has SQL as a file system, doesn’t this muddy the waters a bit as to what an OS named Sequel will be?
And the fact that there is yet another BeOS want-to-be just make it sound like Monty Python’s “Life of Brian” with the “People’s Front of Judea” and the “Judean People’s Front.”
Oh, and by the way, the Yellow Tab Zeta graphic design for the box and etc. just sucks. It in no way makes the OS to be essoteric and simple. Yuck. The whole lack of good design cheapens the package, let alone the ugly BeOS beta 6 interface.
Vic
Did they use their own kernel or is it a modified one?
I thought it was Structured Query Langauge
> feel the lack of “paste as new image”
It will be in a new release, though you can create a new image after you did copy or cut, Refraction will init the new image dialog with the width and height of the image on the clipboard for you. Afterwards you can simply paste.
> Why not put support into yellowTab and their work to
> continue BeOS? Are we going to end up with two seperate
> and competing systems?
YellowTab is a commercial company, not a free effort like OBOS which every one can join. We are not creating another BeOS, so in that perspective we are not trying to compete.
> I have the deepest respect for the people working on
> Sequel and why they need a hobby, but I really can’t see
> why they won’t go with either OpenBeOS or B.E.OS.
Strange how people just don’t seem to get this after the long, long discusions on the OSNews and other forums about this topic.
Most of us have been with OBOS, I have worked on the B.E.OS team. The main difference between them and Sequel is the fact that they are trying to recreate BeOS and we are NOT !
We love the BeOS feel and experience, though we realise that since BeOS R5 is 3 years old now, it will be time to move on, incorporate new techniques and try to improve where Be went wrong (they did in many areas).
> but PLEASE dangit, hardware 3D support is a MUST:)
We do realise this, and there will be 3D support
> Everyone and their mums calls SQL (Sequential Querying
> Language) sequel. Whether its MS or Pervasive or whatever.
SQL has nothing to do with M$, it stands for “Structured Query Language” and just happens to be pronounced the same.
Sequel in our opinion stands for “Follow up” to indicate the fact we want to start where Be left off.
> And the fact that there is yet another BeOS want-to-be
> just make it sound like Monty Python’s “Life of Brian”
Why don’t people just read the whole story before they think they have a wise opinion worth mentioning ?
We are NOT a BeOS want-to-be, we do not aim to be either source nor binary compatible. We embrace a lot of technical designs and the clean API Be Inc. has worked out, so we will model a lot of our OS after these aspects.
> Oh, and by the way, the Yellow Tab Zeta graphic design for
> the box and etc. just sucks. It in no way makes the OS to > be essoteric and simple. Yuck. The whole lack of good
> design cheapens the package, let alone the ugly BeOS beta > 6 interface.
1. Designs are a matter of taste and liking
2. The look of the box has nothing to do with the quality of they OS
3. The BeOS 6 interface has multiple looks including the BeOS R5 one.
4. In stead of picking, propose a better way, or prove you would be the one to say this as you can do better
I always call it S-Q-L.. (i.e. Ess-queue-elle). It’s a very American thing to use the pron. sequel.. though SQL Server has unfortunately undermined this position.
No! It’s squeal!
squeal! squeal! squeal!
And PL/SQL is “peal-squeal”!
Making that MS thing “Squeal Server” which is fairly accurate.
Nice one tpv *L*
Since ‘squeal’ is a monosyllable, MySQL (“mysqueal”), being easier to say, will surely gain marketshare.
PS. Does this mean that “PostgreSQL” is pronounced “Postgresqueal”?
I’m still left with quite a few questions about this interesting effort:
1) Is Sequel open-source and if so is there possibility for sharing of code between any of the other open-source BeOS initiatives (such as OpenBeOS with which I am involved)? If it’s not open-source, what is the procedure for joining the Sequel development team?
2) Did Xentronix write the kernel themselves? Is it based on any other kernels? How is it unique?
3) In what ways will Sequel be *different* from BeOS (and from other OS’s for that matter)? As an OpenBeOS member, I’m very much interested in moving OpenBeOS forward and embracing new technology as soon as possible. If Xentronix has some good ideas, I think that future versions of OpenBeOS could benefit from them too.
4) What types of compatibility (if any) will Sequel have with BeOS? You’ve already mentioned that it will not be API or binary compatible, but will things like drivers be fairly portable between the two? Will the Sequel API’s be BeOSish enough that porting applications will be straightforward?
That’s all for now. Thanks for all of your hard work and good luck with Sequel.
one thing that was probably said but i must have missed, this will be closed source but will it be free or paying?
Will it be x86 only?
What is the longterm commitment of xentronic to the other product that are currently targeted at Zeta?
As a develloper, how my current projects could be bended to fit sequel if i eventually find more advantage to it than current BeOS copycat.
What will the GUI improve on? you mention BFS but will open tracker will be used as well?
Now, for those that wonder why spliting the resource. I learned that 20 programmer is not always faster than 10 programmer. Also i whined for many year that i was stuck with windows on x86, so i will rejoice to see variety in the computer world ecosystem.
The way i see it, Frans want to do glasselevator but do it now instead of building on top of OBOS R1. So the 2 project might merge from a conceptual standpoint in the future anyway.
I respect people who are capable of developing their own OS: but if you want an OS that will be deployed, rather than just to develop it for your own learning and enjoyment, why write an entire OS and userland from scratch? Why not start from Linux, *BSD, etc, and focus on developing the particular features (e.g. pervasive multithreading) that are attractive in BeOS?
(Not thinking about Sequel) there are numerous unfinished projects mentioned on OSnews, with truly amazing goals, run by well-intentioned, highly talented people: many of these projects will never be finished, because the scale of effort required is too high: the 50 engineers for 10 years that someone mentioned above. In the meantime, mainstream computing (projects like Linux, KDE, Gnome) will catch up, because they have more developers, and because after 10 more years of Moore’s Law, the mainstream will be able to do with brute force what BeOS once did with clever tricks.