Cory's Town Hall: Highlights

The short list:
Havoc 2 in 1.7 planned for 2-3 months from now, focus on bug-fixing, HTML-on-a-prim and vector graphics confirmed, Lindens want you to use the voting system for features.

The semi-short list, bolded for emphasis

Cory Linden: "Jillian Callahan: Is Havok 2 on the 1.7 schedule?"
Cory Linden: Yes

Cory Linden: "Francis Chunk: Is the LSL virtual machine moving to Mono? What does that mean?"
Cory Linden: Yes. In the short term, it means that scripts will run at least 100 times faster than they do today. In the longer term, it means that we can allow
scripts to be written in any language that supports the common language infrastructure,
Cory Linden: such as C#, Python, Java, etc. Mono is a 2.0 feature, so Q1 of 2006.

Cory Linden: "Pepplar Sklar: New renderer details?"
Cory Linden: The 2.0 renderer is multithreaded, is designed to scale much better with increased CPU and GPU capabilities, while also being able to scale down
more effectively for low end hardware.
Cory Linden: The biggest change is a move to Open GL Vertex Buffer Objects, which largely eliminate most of the current differences between our ATI, NVIDIA,
and Mac render pipelines.

Cory Linden: "Cristiano Midnight: When do we get an increase in prim limits?"
Cory Linden: A large part of the 1.7 effort is related to scaling and reliability on the sim side and involves fixing aspects of our culling and visibility calculations.
We expect the new approach to be faster which, when combined with Havok 2,
Cory Linden: should allow for an increae in prim limits with 1.7.

Jeska Linden: Hiro Pendragon: When can we expect an API for the GUI so that we can customize it?
Cory Linden: Not in 1.7 but hopefully for 2.0 . . . very high on our list of new features once we are on top of bugs.

Jeska Linden: Icon Serpentine: Is an expanded keyboard control interface in the development plans between now and 2.0?
Cory Linden: yes, but on the 2.0 side of "between now and 2.0"
Cory Linden: next


Jeska Linden: Verbena Pennyfeather: Will LL ever impliment simple programs, like a calculator, web browser, or such into the client?
Cory Linden: yes
...
Hiro Pendragon: Cory, you said HTML / vector will be done... can you answer when?
Louis Neutra: cory u r hot
Cory Linden: not yet, Hiro . . . order will be HTML first, vector later
Hiro Pendragon: HTML gooooood


Jeska Linden: Chavo Raven: when will we be able to write to note cards? and when will we be able to modify object permissions with LSL ? and will we ever get
past the tiny memory allowed for scripts ?
Cory Linden: No, maybe, yes.
Cory Linden: Instead of writing to notecards, we're going to provide a different method for permanent, shared and large storage. Cory Linden: The memory limit will be much easier to get around after mono integration.

Jeska Linden: Hiro Pendragon: When will LL de-centralize the asset server?
Cory Linden: In some ways, we already have, but we'll really have to do it when we have machines in multiple colocation facilities.

Cory Linden: on the 1.7 list is revamped object-object communication so that they won't have to use llemail (which is expensive)

I wish we could write to notecards :/

i can think of reasons it would be nice to write to notecards other than permanent storage.

duh.

Uncle Linden was at my repeater when Cory said notecard writing wasnt' on the menu. Uncle explained that writing to a notecard was actually far more complicated than we would expect. As wonderfully useful as notecard writing would be for other than long-term storage... I guess I'm ok with having some other method for storeage if we're not getting the notecards.

Something tells me that as soon as this other mysterious method of data storge shows up, we may find that reading/writing to it will make sense more then writing notecards, thus abolishing the need to write those automaticly in the first place. =)

Well, save for being able to craft a notecard your script can then can hand to an agent, yeah!

Sure, but if you have the ability to read and write long term storage, your script can hand an agent a scripted object which points to data which you can keep up to date.

neko

Yea good point.. maybe we can start working on a feature suggestion like "llDumpData2Notecard", to transport the data from a certain field into a notecard. =P

Who knows, might be easier since its all server-based.

How much data can be stored with llList2List()? Most of my work involves short lists, I have never capped one. Is the list buffer the same size as a notecard?

Personally, a llList2Notecard(), llString2Notecard() or llDumpData2Notecard() option would be great.

Points to data in what though... a prim with some hovertext? An external website? A landmark they have to teleport to?

Nothing is quite like a notecard. A notecard is just, bam, right there. You open it and it's got up to 64K of text and other assets.

all i know is i want to be able to dynamicaly create and distribute notecards. is that teki wiki enough? how can it be so hard? can't they just have it do whatever SL does when we hit Save on a notecard now? once we can do HTML or RTF or WTF in notecards we need something like ASP or PHP or whatever you use to make templates and dynamic web pages so we can do the same with notecards. writing to a notecard isn't just for storing script info or whatever. that's barely 1% of the uses for writing to a notecard that i can think of. i just used my acronym quota for the year so don't let it be in vain LL.

It's probably a bit more complicated than what meets the eye. Writing to a notecard would require several routines, such as new, append, insert etc. Permissions is probably another can of worms.

I would like to see it sometime soon, it would be very useful.

I would like it to be explained in great detail *WHY* exactly we can't write to notecard, even if I don't understand it. I don't see a reason not to explain why not, since the client should be open source some day eventualy anyway (their words, not mine).

So... WHY NOT, EXACTLY?

Applications other than storage for writetonotecard:

- Instant instructions for something written personalized to a user.

- Output websites to notecard (this is what I want especialy, it would make life much easier for me if I could have all the notecards in the museum auto-update to match the content on the Wiki, as of today, I do this all manualy which takes alot more time).

- Put together a "gift" card from a vendor full of items (highly under used).

- Directions/instructions in a game based on the current location of a user within the game (i.e. Turn to your right to find x).

Probably more that I'm not thinking of.

NoteCards are *very* underused in SL I feel, they can contain anything and can work like a folder yet providing room for better explination for the items. They hold more text than any message you can send giving you the ability to explain ideas better in full before passing them on. They're almost like SL's own version of a webpage, hell you can make a Website with them by having NoteCards within NoteCards acting as page links (something I do with the history records all the time).

So to *not* have NoteCards be dynamicaly editable from LSL scripts just plain sucks. Outright, no arguement, hands down, SUCKS.