My Take on the Firestorm-Phoenix Q&A Video

39:00 – Question: Kid Cyberstar asked… (serious paraphrasing) Will the FS viewer use the new V3 interface and modify it to be Phoenix like?

Jessica explains a bit about what has happened and what has to happen to deal with the new V3 FUI (flexible User Interface). I’ll paraphrase it. Basically FUI is not in FS Release because the Lab kept it secret and did not tell TPV Dev’s what they were doing. The change in the user interface is huge. The change also undoes all the work the FS-PH Team has made building their revised and flexible interface.

The Linden FUI will make it to FS. But, it cannot be directly ported to FS. The team has made too many changes in an effort facilitate the Phoenix style user interface. The problems are; having figure what is or is not compatible, redoing all the incompatible FS interface code, figuring out how make the new stuff look like Phoenix, and the time needed to make the changes.

Realize the FUI in many ways comes close to destroying FS and turning it into a dead end viewer. It is the interface that the Team has been changing. The FS-PH team has a significant amount of work to do to save FS.

Listening to Jessica you will probably realize the team has a lot of revaluating and strategizing to do. They would like to provide interfaces like; Phoenix, V2, Hybrid, and FUI. It may be possible to adopt the FUI and modify it to provide a near Phoenix like look.

42:10 – Question: (didn’t catch name) asked… Will FS be able to upload mesh as well as LL’s viewer? The simple answer is… no. The code used by Linden Lab uses Havok, a third party physics engine that is responsible for keeping you on top of the ground and preventing your walking through walls. Unless a TPV Dev is willing to spring for a rather expensive license for Havok, they will have to reverse engineer a mesh uploader. Since the code is proprietary and licensed by Havok, LL can’t publish the code they use to build the uploader.

One does need to understand that only part of the uploader relies on Havok. It is the part that creates a physics model for the upload. Many modelers build their own physics model because they believe the Havok utility does a poor job. So, while the FS uploader may never be as good as the one LL uses, it may not have to be. As long as it can upload the LoD and Physics models I build, I’ll be happy. I don’t need the Havok reliant part to build my Physics model. So, for me it is very possible TPV’s will be as good as the Lab’s.

43:45 – Question: (I give up on names) Will more PH features port over to FS? Yes. Jessica goes into how they are setting priorities for feature ports and the complications. Feature moving is not a copy paste thing.

46:15 – Question: Would it be a good idea for everyone to boycott mesh so that LL would give it up THEN we could have our Phoenix viewer? O.O

The question and the answer, imo, both miss the point. Jessica supplied an answer in the form of her economic explanation earlier that LL must do away with the  V1 viewers to make money… I just don’t buy it.

Linden Lab has to advance the platform and adopt new technology to survive in the long run. Blue Mars was probably a scare for them. If Blue Mars had handled user created content well, they could have become competition for SL. I suspect most of the new SL users adopt the V3 viewer. Whether some percentage of SL residents stick with V1 has very little effect on where LL is going or what they have to do to compete with other possible platforms. Running the old servers to support V1 viewers is a sunk cost and is probably a small marginal cost to operations.

Also, in regard to the boycott: no one has ever gotten ‘everyone’ to do anything. I personally think the question denotes a very self centered viewpoint… or a good sense of helpful satire… I can’t tell which.

48:00 Break – 50:00 – It was here that I begin to think the video has undergone some editing to trim run time. If so, my thanks to the editor. 🙂

50:00 – Question: What is LL’s response to those using old laptops and computers that cannot support mesh?  (SSE2 is required to run newer viewers V2 and V3 and render mesh. SSE2 was released in 2001 in the Pentium 4.)

Technology moves forward. Older tech gets left behind. Its life, deal with it.

5% of FS-PH run computers that have no SSE2. Those computers will never support mesh. But, the SL platform cannot advance if it tries to support older equipment. If SL does not advance, it won’t survive then no one will have SL to play with. But people just do not believe that. They can’t see it.

The Lab estimates that 2% of SL users are running old computers that do not have SSE2. You guys are going to be left behind. It is up to you to keep up or deal with being left behind.

56:20 – Question: FS and PH seem to use lots of memory. Can that be fixed? Linden Lab is working on it. It is beyond the FS-PH Team to make those fixes as it has to do with the render pipeline.

57:50 – Question: Will the viewer have a feature to save outfits as copies of things or will outfits remain links? No, outfits will remain links. Jessica goes into why. Remember. You can’t copy no-copy items. Links solve that problem. So, repeat after me, “Links are good.”

59:45 – Question: Are labeled buttons coming to Firestorm? They are expecting to do that when they implement FUI in Firestorm. A setting will then control if they are icons, text, or both.

—OK questions were getting boringly repetitious and mostly about various viewer problems and features that are explained in the wiki or can be handled by FS-PH Support.

So, I’m skipping to just the interesting stuff. I’m already at 4,300 words.—

01:13:00 Question: Is LL going back on their open source commitment? The simple answer is no. They are working with open source. Jessica explains why she thinks they are living up to the commitment.

Next page links below the buttons..

14 thoughts on “My Take on the Firestorm-Phoenix Q&A Video

  1. Good “summation” Nalates. I too watched the whole thing and you seem to have hit the important points.
    One off-hand comment Jessica made about Viewer1 caught my attention; right now LL is running 2 versions of the inventory loading program on the servers. When the day comes that they remove the old one, V1 will still “work” but without your inventory.
    People need to bite the bullet and learn to use a V2 viewer. They are, after all not still running Windows 97. Changing SL viewer is far less traumatic than changing OS 🙂

    • Thank you. 🙂

      Thinking of two inventory systems can be misleading. Jessica usually doesn’t go too techie. But, the lack of technical understanding by residents is leading to some of the frustration and spin, imo.

      Inventory is a list of items that reside in a database. The list itself is data in a database too. Having two inventory systems is not a matter of having two databases. It is having two API’s (Application Programming Interface). The servers that handout asset information (e.g., the stuff that makes up a sexy top) have new coding to handle newer faster ways to find and handout asset information. There is new code to process the inventory lists too. That new code has a function name that is somewhat like a phone number, which makes up the API. We can see functions in the Linden Scripting Language (LSL) to see what that looks like. They look like: llSetPrimParams( [instructions here]). We now have a new one llSetPrimParamsFast([similar but diff instructions here]).

      Basically your viewer is programmed (by using either llSPP() or llSPPF() ) to call either the new code or the old code. Eventually the old code will be removed from the system and no longer work. It is kinda geeky semantics and I think irrelevant to most of us whether the new and old code live in the same or different servers. Also, how one thinks of the meaning of ‘system’ in this context can shift one’s meaning.

      Talking with Oz Linden in the Open Source Meeting he pointed out that all the new API’s have been made available to TPV Dev’s. So, it is possible for the TPV Dev’s working with V1 to update their viewers and keep them working. How difficult that will be I don’t know.

      The Lab will not be updating SLV 1.23.x. That viewer is doomed. They think making the changes to keep it working is too much work. They decided to focus on new things. Linden Lab is NOT focused on killing V1 viewers. They don’t care about V1’s, as best I can tell. They are abandoning THEIR 1.23 viewer. They are still providing lots of information so others can maintain the TPV1’s.

      Jessica sees a business need for the Lab to kill off all V1’s. I don’t. I think the Lab is just moving on and V1’s have very little financial impact on LL.

      In a nuanced situation with many complexities, Jessica is pointing at the Lab as the reason Phoenix is going to be discontinued. The FS-PH Team may truly perceive the situation that way. I don’t. I think the FS-Ph Team is making the same decision the Lab made. I think they have made it clear they think working on Phoenix is too much trouble, counterproductive, and they want to move on to the new stuff.

      If I were making free viewers, I would make whatever I wanted to make and use personally, which may be what Milkshake is. I would probably provide free copies to friends. Whether anyone else wanted to use it or not, would not matter to me. I would have no reason to care. Anyone that yelled at me about MY viewer would be blocked from downloads. I don’t want the aggravation.

  2. Wow. Excellent post. I haven’t seen the relevant video (and have no intention of wasting my time on it as the LL viewer works fine and I don’t bother with others) but I’ve heard about it of course. Personally my reaction would be the same as your final comment; all whiners would just be blocked, not appeased.

    There is a lot of childish, uninformed behavior in SL (for example, people screaming at someone who is using particles — one of the most useful and beautiful things in SL — telling them to “quit lagging me!” when all they have to do is turn down the particle count in their own viewer), but in my opinion the so-called “viewer wars” are the most assinine. ALL software morphs over time as the programmers gain more knowledge/experience and the technology improves. The obsession with an old interface and refusal to move forward is laughable, and the vitriol directed at the official LL viewer is ridiculous. Personally I think LL should make the use of their viewer mandatory. The use of other viewers dilutes the input stream to the programmers of the LL viewer and it wastes the time of support staff being hammered at to fix problems that are 3rd-party viewer related and not their concern.

    And anyone who had the thought “If they take away my [insert name of TPV here] then I’m leaving SL!! I’ll show THEM!!” please see my remark regarding childish behavior above.

    • Unfortunately the behavior is not limited to SL…

      Snap judgments and jumping to conclusions are typical human behavior. Add to that the entitlement mentality now being encouraged by many political systems is filtering into other parts of life. So, it is becoming your duty to stop using particles because they can’t be bothered to change their settings or have yet to figure out what is going on.

      Enough whining on my part….

      As to other viewers… they do have some fun stuff in them. The Linden Lab viewer is sort of the mid-range Ford Explorer. TPV’s range from a little Volkswagon or a work truck to a Ferrari. Or may be a gadget rich Volvo.

      TPV’s are not about ‘need’ so much a fun. Some of the first TPV’s were identical to the SL Viewer but with more bug fixes, they just worked better.

    • I watched the full video and I only use the official viewer. Its still a good watch. Very informative on goings on and kinda shows how the whole ‘FIC’ thing really works (some people get into the favored crowd merely because they show up when others don’t). 🙂

      It gave me a lot more sympathy for the Firestorm / Phoenix devs than I had before. 🙂

      Some of the questions were downright irrational. Some people asked the same question others had asked many times already, insisting no one had asked it and that no answer had been given – demonstrating that they really don’t want an answer, they just want to vent and blame. Which is sad.

      People are very upset at their cheese being moved. The rational option is to learn how to handle the new location of one’s cheese, because it won’t move back. But they’re being emotional instead – as if anger would cause their cheese to come back to them.

  3. You can bad mouth Cool Viewer and v1 viewers all you want, but as long as they continue to get higher frame rates and better performance then v2 and v3 viewers then people will keep using them. Perhaps LL should spend a short period of time on code efficiency in order to bring v2 and v3 up to standards.

    • I’m not seeing where anyone is bad mouthing Cool Viewer. Henri has made a good viewer and is showing that V1’s can continue to be upgraded and work.

      Also your ‘should’ shows you are not aware that is what the Lab has been working on for the last few weeks. The Shining Project has been optimizing and fixing problems in the new render pipeline and OpenGL compatibility. The latest 3.2.6 is quite a bit faster.

  4. Thanks for your post Nalates; English is my second (rusty) language, it would take me two days decode the two hour video, your summary and your good judgment is a great help.

    Last year i was a reticent Emerald and official v1.23 viewer user. I was comfortable with those viewers and i didnt want to change.. but some day i decided test v2.3 (or something near). I dont remember exactly what version it was, but was the faster viewer i ever had. I could have sharders activated even in cowled placed (my system have like 10 years old today, so that was a great improvement to me). Due to economics reason i am still stucked to this oldie PC, i had to learn get along good with lag and less than 10 fps everywhere, but furthermore since v3.0 i freeze like crazy in every new viewer out there.. its unusable for me so i had to return back to v1, Singularity is my viewer now. I miss a lot of things of the last releases, but this is the only way i can logging SL today. Hopefully, i expect have a new computer in few months, but now i feel a little relieved knowing that LL is not interested in turn off v1 platform in short. I’m grateful with every developer out there for help grow our little universe, and i am completely at the mercy of Henri in particular for let me enjoy the magic of mesh these days. Kirstenlee departure is for sure a sad notice, and even though couldnt install her viewer, i feel that a part of SL go with her.

    @Caliburn: On my point of view make the official viewer the only option is a very bad idea. To me they are not only a fun.. but a useful think to everybody, even if you are not gonna use them. TPV developers help to fix an important amount of common bugs (and they have their own jira system to fix the bugs of their viewers), give useful feedback to LL and release new features that could be in the official viewer.. or not: and i am not talking just about the “Show Look At” that everybody feel like essential… Ligthing and Shadows, RLV, RLVa.. the nice Collapse button that i always miss on my official viewers!…
    Its nice if you like official viewer.. really.. i would use the developer viewer if i could, and i completely agree when you refer to the those intransigent users that all they do is complaint about everything.. after more than year reading people growling about UIs, i think that if i hear someone telling that FUI sucks my head will explode in hundred of tiny pieces of slimy brain.

    I would say that its a bit childish of your part too complaint about people using TPVs. The more options the better and the richer SL is, because what works for you maybe dont works for me, because what you like maybe is not of my taste.

    About the particles matter… yeah its very pretty, i use them often in my creations, but being a mentor on a sandbox i can tell you that it can be very very very disturbing. Certainly, i know the shortcut to turn it off when it disturb me, but i am responsible of keep an appropriate atmosphere in order to build in peace, and sometimes (always) is better and faster ask politely turn off the particles emiter to the resident/griefer than try teach every newbie in the sandbox on to turn it off in their Preferences… and you know… maybe what you like could be not what i like or what other person likes, we live in a community ^^

  5. Thanks for that summation Nalates – and sitting through 2 and a half hours for the rest of us:). I do agree with your thoughts on part of the Phoenix user-base – I think they inherited them from the Emerald days.

    Incidentally Cinder does not believe Milkshake is ready for wider release (just ask her) – it was created for herself and a few friends, though hopefully she may develop it further in the future.

  6. I’m getting tired about the “v1 vs v2” stuff and people who are not even true programmers and pretend that “v1” TPVs will die or be soon unable to function properly on the grid. Like I wrote in details on my forum (http://sldev.free.fr/forum/viewtopic.php?f=5&t=584), “v1” viewers are not v1 any more (all the underlying ll* libraries, most major viewer classes and even the whole rendering pipeline are in fact v2/3 ones in the Cool VL Viewer, Singularity and now even Phoenix since it reused all my patches to implement mesh).

    No, “v1” TPVs (and more exactly TPVs with teh v1 UI) will NOT die any time soon ! Stop the FUD !

    • Thanks Henri. It is hard for people to know what developers are doing without reading through the development repositories. Also, without understanding how the viewer code is constructed, it is hard to read about viewers and take away the correct impressions and nuances.

      The idea many of us take away when a developer says their viewer is based on V1 or Snowglobe code is that new code is being changed to work with older code. If the old code is being modified to work with new code, then I suspect few of us are thinking in that frame.

      I keep expecting a developer to break the viewer render engine and user interface into separate modules. I also expect the viewer code to become more and more modular. But, we don’t hear about those details.

      I would think someone would be developing an app for SL to run on tablets and smart phones. I suspect that development would be kept secret by LL and anyone else developing it. But, the financial rewards could be enormous. With 400,000 downloads of FS/PH I would assume an app would sell maybe 200,000 apps. While it would probably require a subscription to go with the app, people are thinking along that line as Hamlet’s poll shows.

  7. One thing you should know Nalates:
    There is a *free* implementation of the Havok Physics engine SDK available for those who want to develop games which cost $10 or less.

    Read here: http://www.havok.com/sales

    • There seems to be a gotcha in the licensing details. For development of a game there is no cost. As long as the game distributed for less than US$10 there is no license required to distribute it.

      A viewer is only part of the SL ‘game’ that has a premium retail value of US$9.95/month or about a $120/year or $72 per year on an annual basis. It becomes a point of interpretation of the Havok license as to whether a viewer qualities for the free, basic, or pro license. Since most TPV Dev’s are much more into the licensing than I am and are not using the free Havok, I’m guessing the license is in fact a problem. The same can be said for OpenSim.

      • I think you have it backwards, Nalates. As a sometimes-dev (not of viewers), if I have to care about the license, I won’t use a thing for a personal project. I leave it to my employers’ lawyers when its and issue for them. I’d avoid Havok, because I couldn’t afford the lawsuit if I’m wrong and am not lawyer enough to be confident I’m right.

Leave a Reply to Jack Abraham Cancel reply

Your email address will not be published. Required fields are marked *