atomos-discuss Mailing List for Atomic OS
Status: Pre-Alpha
Brought to you by:
dleet
Menu
▾
▴
atomos-discuss — General discussion list about Atomic OS
You can subscribe to this list here.
| 2006 |
Jan
|
Feb
|
Mar
|
Apr
|
May
|
Jun
|
Jul
(1) |
Aug
|
Sep
(23) |
Oct
(4) |
Nov
|
Dec
(1) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 2007 |
Jan
(15) |
Feb
|
Mar
|
Apr
|
May
|
Jun
(1) |
Jul
(1) |
Aug
(1) |
Sep
(1) |
Oct
|
Nov
|
Dec
|
|
From: Scott E. <ps...@gm...> - 2007-09-02 02:16:32
|
I've put up a new poll at psema4.com. I know a Beta release is still a ways off, but I'd like to get an idea of what people would like to see well before hand. Which tutorial application would you like to see accompany a beta release of Atomic OS 1.0? * Address Book * Bookmark (Hyperlink) Manager * Notebook * Presentation Generator * Other Any thoughts or suggestions would be great. -- Scott Elcomb http://www.psema4.com/ |
|
From: Scott E. <ps...@gm...> - 2007-08-31 20:39:06
|
Hi all, Yesterday, while researching Web OS related topics, I came across the GNUCitizen[1] site. The site focuses primarily on security issues and after some browsing around, I came across an interesting little contest they're running - with the winners receiving some exposure on their site and a book entitled "Cross Site Scripting Attacks: Xss Exploits and Defense."[2] The description of the contest[3] is: "For the month of August we are going to collect various pieces of work that inspire the hacker spirit. This means that we except[4] projects that are creative, clever, unique, provocative, intelligent, intense, intriguing or simply interesting." Unfortunately, the contest ends today. If you think that Atomic OS[5] is a worthwhile entry, please send an email (saying why) to contact-at-gnucitizen.org and group-at-gnucitizen.org Thanks in advance - I know it's just a book, but it is a book that could help secure AOS. ;-) Take care all, - Scott [1] http://www.gnucitizen.org/blog/web-os [2] http://www.amazon.com/Cross-Site-Scripting-Attacks-Exploits/dp/1597491543 [3] http://www.gnucitizen.org/projects/the-month-of-hacker-folklore [4] I assume this was supposed to be 'accept.' [5] http://www.psema4.com/ -- Scott Elcomb [ http://www.psema4.com/ ] "Our Founders' faith in the viability of representative democracy rested on their trust in the wisdom of a well-informed citizenry, their ingenious design for checks and balances, and their belief that the rule of reason is the natural sovereign of a free people." - Al Gore (The Assault on Reason, 2008) "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Scott E. <ps...@gm...> - 2007-07-18 20:13:51
|
Here is the first draft of the Atomic OS Manifesto, as posted on www.psema4.com with minor modifications (sections are numbered for clarity in text-only format). Comments and thoughts would be appreciated. TIA! - Scott The Atomic OS Manifesto 1.0 - What is AOS or Atomic OS? In a nutshell, "Atomic OS is a responsive Web 2.0 operating environment & development platform. Based on AJAX techniques, it emulates/provides standard operating system features including a command-line shell, interpreter, filesystem, database access and GUI services." 2.0 - What it isn't! Atomic OS is not a program. By itself, Atomic OS isn't capable of doing anything. Developers use AOS as a library to make web-based applications. Atomic OS is not an Operating System! Instead, it is a library that, when combined with custom JavaScript code and executed within Mozilla Firefox, simulates an OS-like environment that web developers can leverage. The base AOS "distribution" comes with a set of tools (such as the WASH shell) that allow developers to interactively explore Atomic OS and it's features. This doesn't make it an OS. A desktop-interface / window-manager does not change things either. The reference to "Operating System" in the project's name is nothing more than a convenient metaphor. 3.0 - Why write it? As a web administrator (/master/developer) I was frequently frustrated by the limitations imposed by browser and WWW programming technologies. Creating intuitive (and responsive) User Interfaces for end-users was very difficult. DHTML (AJAX without the server interaction) applications are capable, but they can become very complex very quickly. Atomic OS aims to make things a little more clear by following the examples of data organization (file systems) and user interfaces (shells and desktops) used by virtually every modern Operating System. The complete AOS library, the filesystem, supporting tools and utilities, custom application(s), and end-users' data are all stored within one HTML file =96 providing an extremely flexible solution to several difficult issues I've faced with developing web-based applications. 4.0 - Why everyone will benefit. Programmers benefit from the OS-like API. Programmers may also find the embedded development tools useful. Although preferred, it is not necessary to "open-source your applications." The GNU LGPL license allows for linking to non-free software; this feature may be useful for some businesses. Administrators benefit from enhanced organization, self-containment of run-time data and logs, and improved management interfaces. End-users benefit from lower learning curves. By mimicking the look and feel of modern desktop Operating Systems and applications, users find themselves in a familiar and intuitive environment. 5.0 - How will Atomic OS be available? Atomic OS is available from SourceForge. Please visit http://sourceforge.net/projects/atomos for more information and downloads. 6.0 - How you can contribute. If you are a JavaScript programmer, consider helping out by porting GNU utilities. If you are a graphics designer, we could use your help in creating desktops and themes. If you are a technical writer, we could really use more concise documentati= on. Any monetary donations will be put towards paying developers for their time and efforts. Depending on resource availability and circumstances, such funding may also be used towards creating and/or maintaining an Atomic OS Foundation. Please send inquiries, broken links, other corrections and/or suggestions to psema4@... Copyright (c) 2007 Scott Elcomb Revision: 0.1 - 07/18/2007 --=20 Scott Elcomb [ http://www.psema4.com/ ] "Our Founders' faith in the viability of representative democracy rested on their trust in the wisdom of a well-informed citizenry, their ingenious design for checks and balances, and their belief that the rule of reason is the natural sovereign of a free people." - Al Gore (The Assault on Reason, 2008) "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Scott E. <ps...@gm...> - 2007-06-14 18:22:23
|
Hi all, For those who're still subscribed to this list, my apologies on the long delay since last I posted. I'm still battling an insidious illness. However, it's still a battle so here's my next volley. =) I posted this message to the Toronto Linux Users Group (previously TLUG, now GTALUG) earlier today. Would love to hear any comments. Take care all! - Scott ---------- Forwarded message ---------- From: Scott Elcomb <ps...@gm...> Date: Jun 14, 2007 1:59 PM Subject: [TLUG]: [OT]: Seeking general Atomic OS usability comments To: GTALUG <tl...@ss...> Atomic OS is first and foremost a web development library. However since it is also the operating environment for it's applications, I'd be curious to hear what people think of it in general. There is a complete re-write of Atomic OS / WAJAX at http://projects.psema4.com/atomos/ This version is almost entirely object based, and, unlike previous prototypes, is intended to become a solid foundation for future versions. It is expected to become version 1.0 upon release. If anyone would like to take a look and make comments, it'd be most appreciated. Before visiting there are a couple points that should be noted: - Only firefox is supported. - Either open a second Console after browsing to the page, or 'cat /etc/motd' for a list of important notices about this version. - Very few commands exist. It basically provides a shell and browser. What I'm looking for is whether or not it is *close* to intuitive for linux users. There's still a long way to go on this project, but many of the problems defined by previous protoypes cores have now been dealt with. TIA. - Scott. -- [..sig..] -- The Toronto Linux Users Group. Meetings: http://gtalug.org/ TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists -- Scott Elcomb "Our Founders' faith in the viability of representative democracy rested on their trust in the wisdom of a well-informed citizenry, their ingenious design for checks and balances, and their belief that the rule of reason is the natural sovereign of a free people." - Al Gore (The Assault on Reason, 2008) "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Weirdbro <wie...@gm...> - 2007-01-12 10:32:04
|
I am familiar with the canvas tag. It has really slow pixel rendering, so I found that making it work like a monitor, and least a raster monitor, is impossible. On 1/11/07, Scott Elcomb <ps...@gm...> wrote: > > On 1/11/07, Bryan Elliott <for...@gm...> wrote: > > Scott: > > > Nice clean code. I love it, and the dock in particular. Has me > > > wondering if a version could be implemented for the AOS desktop. > > > Along the lines of an applet maybe? > > > > Perhaps, but note that performance decreases dramatically with a greater > > number of icons. There's little that can be done about it; it's because > of > > the way resizes are done in JavaScript under a browser (ie: somewhat > > inefficiently). I mean, one could theoretically write Warcraft in > > Javascript. All the components to do so are there and available (ie: > sprite > > placement and control), but (and I know this from experience; I > tried. Had > > a good pathfinder when it was just two peons and a few rectangular > barriers. > > ) when you get up to sheer numbers of stuff (at least 256 just for a > 16x16 > > viewport), performance drops down the tube, into the septic tank, and > keeps > > on racing down until suddenly bursting out of the ocean just off the > coast > > of Australia. > > Ok, please forgive me if this post isn't entirely coherent... Just > finished a 17 hour shift. Lol - one of the benefits of being on-call > 24/7. ;-) > > I can understand your comments about applying game development theory > to web design... In fact, that might even be a good way to throttle > the performance issue. > > Say that performance is acceptable with 6 icons, but not 16. Why not > use parallax scrolling techniques? That's the way old 2D side > scrollers (like Super Mario World for example) did it IIRC. > > Use an array that contains indexes to the icons, and then use a > pointer and "window width" to show only the 6 icons currently > available + portions of the two icons on either end of the "window" > that may be (potentially partially) visible. CSS clipping techniques > should work for this, same as it does for DirectX, SDL, OpenGL, > Allegro, and other Graphics/Game development libraries. > > Using this technique the browser won't have to render/scale the 10 > "extra" images all the time, therefore keeping performance manageable. > (Not sure what impact preloading the icons would have in this > situation but I suspect it would help... maybe not during page > loading, but at "execution-time" perhaps.) > > I have several game programming texts laying around and would happy to > provide some pseudo-code if you'd like. > > K, need some sleep. I'll reply to your other post just as soon as I > get up and am more than 20% concious. ;-) > > - Scott > > PS - Are you familiar with the canvas tag? <insert-evil-grin-here> > > -- > Scott Elcomb > http://atomos.sourceforge.net/ > http://search.cpan.org/~selcomb/SAL-3.03/ > http://psema4.googlepages.com/ > > "They that can give up essential liberty to obtain a little temporary > safety deserve neither liberty nor safety." > > - Benjamin Franklin > > '"A lie can travel halfway around the world while the truth is putting > on its shoes." > > - Mark Twain > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Atomos-discuss mailing list > Ato...@li... > https://lists.sourceforge.net/lists/listinfo/atomos-discuss > |
|
From: Bryan E. <for...@gm...> - 2007-01-12 03:26:08
|
> Say that performance is acceptable with 6 icons, but not 16. Why not > use parallax scrolling techniques? That's the way old 2D side > scrollers (like Super Mario World for example) did it IIRC. Well, for one, I don't understand how parallaxing applies; it's the trick of using two or more repeatable backgrounds, all but the deepest of which is transparent, to create a 3d-like effect by scrolling them at different rates. This could be accomplished in Javascript relatively easily, using png's and specialized behavior for IE. Still, it doesn't really apply to a dock. That's jsDock Mk IV I've got up. I did get a performance gain (in the move from III to IV) by deflowing the elements (ie: positioning them manually, rather than making the browser reflow them), but not much; the limiting factor is ultimately the internal resize function for a given browser. One could, perhaps, tweak firefox to render all images as GL objects, thus moving the processing over to the user's video card, but that's way beyond my own scope (but would dramatically increase the performance of most image drawing capabilities of Firefox). > Use an array that contains indexes to the icons, and then use a > pointer and "window width" to show only the 6 icons currently > available + portions of the two icons on either end of the "window" > that may be (potentially partially) visible. CSS clipping techniques > should work for this, same as it does for DirectX, SDL, OpenGL, > Allegro, and other Graphics/Game development libraries. That's how I did things in my Warcraft attempt; large images of sprite-parts, arranged so that the code would already 'know' where each sprite would be. Believe me when I say that not only is it not faster (especially for anything with alphas; 'Fog of War' takes a whole second to render on a 16x16 grid), but you can't scale backgrounds in CSS, so you lose the niftiness of a zooming dock. Though, having a 'window' that is smaller than the combined widths of the icons, and positioning the icons so that, relatively, the one you're over is the one that's under your cursor (it's an annoying function to debug, but the way it works is that, say, for a sum width of N, and 16 icons, the area in which icon 0, for example, is under your cursor is N/16 pixels wide). Not to mention that it undermines the 'modularity' concept behind the design of the jsDock; the designer should be able to simply include images of any sort, and have them behave like an icon. Though, I gotta say that the true niftiness of the WC attempt was figuring out how to dynamically load in the components of a new unit (ie: the JSON for its controls, the actual code for its AI, the CSS for its movement, stylization, special powers, etc). > Using this technique the browser won't have to render/scale the 10 > "extra" images all the time, therefore keeping performance manageable. > (Not sure what impact preloading the icons would have in this > situation but I suspect it would help... maybe not during page > loading, but at "execution-time" perhaps.) Preloading wouldn't do anything; they don't get reloaded, only rescaled. Add to it that the icons are all scaled at different sizes, and that, again, you can't scale backgrounds, it wouldn't work. Truth is, css-backgrounds actually perform WORSE than straight up images. (create and deposit X number of normal images, timing it. Now do the same with the same number of DIVs given a class what makes it emulate an image via the background-image property. Generally (at least in Firefox), the former takes less time than the latter). > I have several game programming texts laying around and would happy to > provide some pseudo-code if you'd like. I'm pretty good with game programming; did a few chintzy ones back in my old VGA programming days (of note: when I was 16, I made a version of Snake that would go all warbly and drunk-like the more 'poison' apples you'd eat, and more blurry the more 'drunken' apples you'd eat. Never did quite get the memory leaks out of it, but I lost it to my first bad experience installing Linux, so...) > PS - Are you familiar with the canvas tag? <insert-evil-grin-here> No, but (Wikipedias it) that looks nifty! -- Bryan Elliott for...@gm... http://www.fordi.org |
|
From: Coaster <rol...@gm...> - 2007-01-12 02:46:48
|
Hi, I'm a spectator here... (Friend of Weirdbro's.) Why use HTML? Given the fact that you aren't marking up hypertext, a custom XML definition would probably be better suited for the task. (Just my two cents.) On 1/11/07, Bryan Elliott <for...@gm...> wrote: > Application GUIs should be built in HTML as groupings of DIV elements, and > such, with CSS for positioning and stylization, but not directly included > (ie: include the Window as a div style='display: none', and copy its > contents out via DOM to create an instance). Well, that would be the best > way, but it doesn't always work, so normally we'll just toss it into an > innerHTML, link in the behaviors, and pray for workingness ^_^. > > An unfortunate thing about IE is that it doesn't play well with 'custom' > tags. For example, <window>Some innerHTML</window> confuses IE to the point > that the tag gets registered with the DOM, but not it's innerHTML - and the > ending tag throws an error. As a result, our predefined UI tags should be > of the form <div class='window'></div> > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > _______________________________________________ > Atomos-discuss mailing list > Ato...@li... > https://lists.sourceforge.net/lists/listinfo/atomos-discuss > > > |
|
From: Scott E. <ps...@gm...> - 2007-01-11 21:22:48
|
On 1/11/07, Bryan Elliott <for...@gm...> wrote: > Scott: > > Nice clean code. I love it, and the dock in particular. Has me > > wondering if a version could be implemented for the AOS desktop. > > Along the lines of an applet maybe? > > Perhaps, but note that performance decreases dramatically with a greater > number of icons. There's little that can be done about it; it's because of > the way resizes are done in JavaScript under a browser (ie: somewhat > inefficiently). I mean, one could theoretically write Warcraft in > Javascript. All the components to do so are there and available (ie: sprite > placement and control), but (and I know this from experience; I tried. Had > a good pathfinder when it was just two peons and a few rectangular barriers. > ) when you get up to sheer numbers of stuff (at least 256 just for a 16x16 > viewport), performance drops down the tube, into the septic tank, and keeps > on racing down until suddenly bursting out of the ocean just off the coast > of Australia. Ok, please forgive me if this post isn't entirely coherent... Just finished a 17 hour shift. Lol - one of the benefits of being on-call 24/7. ;-) I can understand your comments about applying game development theory to web design... In fact, that might even be a good way to throttle the performance issue. Say that performance is acceptable with 6 icons, but not 16. Why not use parallax scrolling techniques? That's the way old 2D side scrollers (like Super Mario World for example) did it IIRC. Use an array that contains indexes to the icons, and then use a pointer and "window width" to show only the 6 icons currently available + portions of the two icons on either end of the "window" that may be (potentially partially) visible. CSS clipping techniques should work for this, same as it does for DirectX, SDL, OpenGL, Allegro, and other Graphics/Game development libraries. Using this technique the browser won't have to render/scale the 10 "extra" images all the time, therefore keeping performance manageable. (Not sure what impact preloading the icons would have in this situation but I suspect it would help... maybe not during page loading, but at "execution-time" perhaps.) I have several game programming texts laying around and would happy to provide some pseudo-code if you'd like. K, need some sleep. I'll reply to your other post just as soon as I get up and am more than 20% concious. ;-) - Scott PS - Are you familiar with the canvas tag? <insert-evil-grin-here> -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Bryan E. <for...@gm...> - 2007-01-11 19:00:04
|
Scott: > Nice clean code. I love it, and the dock in particular. Has me > wondering if a version could be implemented for the AOS desktop. > Along the lines of an applet maybe? Perhaps, but note that performance decreases dramatically with a greater number of icons. There's little that can be done about it; it's because of the way resizes are done in JavaScript under a browser (ie: somewhat inefficiently). I mean, one could theoretically write Warcraft in Javascript. All the components to do so are there and available (ie: sprite placement and control), but (and I know this from experience; I tried. Had a good pathfinder when it was just two peons and a few rectangular barriers. ) when you get up to sheer numbers of stuff (at least 256 just for a 16x16 viewport), performance drops down the tube, into the septic tank, and keeps on racing down until suddenly bursting out of the ocean just off the coast of Australia. -- Bryan Elliott for...@gm... http://www.fordi.org |
|
From: Bryan E. <for...@gm...> - 2007-01-11 18:52:56
|
Weirdbro: > Bryan, thats some nice JS code. I can't do anything nearly that useful. I > did do some metaprogramming to get the Process system working, and thats > probably the best JS code you can find for me. Ooh, can I see? Processes, locks and mutexing are something I'm only peripherally familiar with, and they seem like interesting work, espeically in a lax environment like JS. Link me? Scott: > I'm familiar with the quirksmode site. Lol. 'Been visiting it > on-and-off for years. Definately good to have a x-browser developer > on board who's familar with some (of the seemingly zillions of) > browser quirks. There's really not that many; it really just comes from knowing how different implementations of a given object/function/element/style will behave, and you only really notice them when a page acts all wonky on a given piece of work. By and large, things work well together; the trip-ups are few and far between, but just common enough to be irritating. Scott: > How do you see the seperation of the project components? I'd like to > delve into this topic deeper - it could have a serious impact on the > way the project is developed. I've had some difficulty organizing and > documenting; suggestions would be more than welcome. Well, there's the standard application model for development, except for one thing: there's also a client-server comms model that should be handling a bunch of stuff. Essentially, there should be a base API with a number of sectioned-off functions, like windowing, file control and the like. There should also be an asynchronous communication (AC) library for talking to the server. The API should probably also have an IPC method for communicating with other applications the user has permission to talk to (not, mind you, necessarily in the same context; it should be able to use the AC lib to have the server talk by proxy to clients running in different contexts). Additionally, there should be a simple metaphor to handle things like finding out what devices are connected to the remote system. Determining how to play sound or video, as another example, should be transparent to the application writer (ie: locally if the terminal is the local computer, via Quicktime and streaming if it isn't and the client is firefox, via an embed if it's IEx, etc.). I'm thinking on how to make the event model operate uniformly as well, but give me time and experimentation, I'll get it done. Essentially, if we want anyone other than ourselves building applications for this system, we have to make the programming process both easy and painless. Also, we need to have certain 'base' widgets, such as: Your normal window widgetry, like labels, text inputs, timers, etc. HTML handles most of these by itself, so little work there. Things like an HTML frame (with callbacks), a rich editor, etc. An OS is less than a shell without some standard interfaces. I'm thinking most of the API could be built into a single object (Application, or something similarly innocuous). Because of the nature of Objects in Javascript, it can be built over the course of several different files, each covering a different object, and thus, a different aspect of the API. Documentation shouldn't be hard, just tedious; implement each method and document what it does and where it could be used. Application GUIs should be built in HTML as groupings of DIV elements, and such, with CSS for positioning and stylization, but not directly included (ie: include the Window as a div style='display: none', and copy its contents out via DOM to create an instance). Well, that would be the best way, but it doesn't always work, so normally we'll just toss it into an innerHTML, link in the behaviors, and pray for workingness ^_^. An unfortunate thing about IE is that it doesn't play well with 'custom' tags. For example, <window>Some innerHTML</window> confuses IE to the point that the tag gets registered with the DOM, but not it's innerHTML - and the ending tag throws an error. As a result, our predefined UI tags should be of the form <div class='window'></div> |
|
From: Scott E. <ps...@gm...> - 2007-01-10 04:10:27
|
On 1/7/07, Bryan Elliott <for...@gm...> wrote: > Danks ^_^. I have some ideas for a mimimalist environment within which > AtomOS could work. S'why I was asking about the backend stuff before. Glad to hear that. I've played with creating customized distros, and while it truly is fun and very educational, it's also quite time consuming. Love to hear what you think such an environment should look like. I've been thinking lately to use the LFS project (http://www.linuxfromscratch.org/) as a base for the image though I'm not sure it'd be worth the effort just yet. > > Would you be interested in taking on an AOS-specific > > customized live CD? Would you like to hack on the AOS code? > > Sure, on both counts. I have a bit of experience hacking on things like the > backend bash daemons in Slax and Ubuntu that handle disk recognition, > process throttling, etc (could be better done in Perl or C, but hey; not > everyone does perl, and writing it in a compiled lingo means distributing > the source). > > Also, I'm almost as browser quirks fluent as this guy (though I'm not a very > good writer, so no book to my credit), with one exception: I'm better at > XMLHttpRequest / iFrame / Script include wrangling (ie: I found out good > ways to fall back to other techniques if XMLHttp isn't working and we're > using JSON). So, I'd really like to have a stab at debugging cross-browser > compatibility issues. Absolutely wonderful. =) I'm familiar with the quirksmode site. Lol. 'Been visiting it on-and-off for years. Definately good to have a x-browser developer on board who's familar with some (of the seemingly zillions of) browser quirks. > I'd also like to have a hand in separation of design / presentation / > structure / content / behavior of the OS (I've become pretty practiced in > it, so far). How do you see the seperation of the project components? I'd like to delve into this topic deeper - it could have a serious impact on the way the project is developed. I've had some difficulty organizing and documenting; suggestions would be more than welcome. > Oh, if anyone wants to see some stuff I've done recently: > http://wwwdev.wharton.upenn.edu/alumni (what I do at my > real job - on the development server 'cos it won't be live 'till monday. > Take a look at the JS for the navmenu and the tabbed boxes; they're great > examples of how I do JS OOP). > > http://www.fordi.org/comics (the neat little dock at the bottom is my work; > the pages are stuff I like to look at. For some reason, I'm not fond of > bookmarks. I'm actually working on it; a '+' symbol to add a link, and a > nice interface for choosing the image) Nice clean code. I love it, and the dock in particular. Has me wondering if a version could be implemented for the AOS desktop. Along the lines of an applet maybe? -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Scott E. <ps...@gm...> - 2007-01-10 03:39:28
|
On 1/8/07, Weirdbro <wie...@gm...> wrote: > Bryan, thats some nice JS code. I can't do anything nearly that useful. I > did do some metaprogramming to get the Process system working, and thats > probably the best JS code you can find for me. I'd suprised if that remains true for long - you took me off gaurd a few times with your JavaScript. I'm not that great at JS coding yet, but your solution for creating processes was elegant and done extremely quickly. Quite literally no-small-feat. =) I'm still torn as to whether or not to include that code. I'd really, really like to since it'd be in keeping *nix style system/software design. As you pointed out it though, it might not be the best thing for AOS to have. Particularly since we don't really need it atm. (Incidentally, the same is true of the combined I/O terminal you and your friend came up with. Very cool stuff that I'd love to see AOS using eventually.) Your Pure-JS build system and OOP kernel changes will be integrated with the 0.1.0 release I'm planning to get out this month. There's no need to be modest. ;-) -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Weirdbro <wie...@gm...> - 2007-01-08 23:51:10
|
Bryan, thats some nice JS code. I can't do anything nearly that useful. I did do some metaprogramming to get the Process system working, and thats probably the best JS code you can find for me. On 1/7/07, Bryan Elliott <for...@gm...> wrote: > > > Naturally we need JS hackers, but there's also some other critical > > items to be looked at. You have some history with SLAX and were the > > author of MyDSL Maker (Personally my single-most-admired web-based > > tool). > > Danks ^_^. I have some ideas for a mimimalist environment within which > AtomOS could work. S'why I was asking about the backend stuff before. > > > Would you be interested in taking on an AOS-specific > > customized live CD? Would you like to hack on the AOS code? > > Sure, on both counts. I have a bit of experience hacking on things like > the backend bash daemons in Slax and Ubuntu that handle disk recognition, > process throttling, etc (could be better done in Perl or C, but hey; not > everyone does perl, and writing it in a compiled lingo means distributing > the source). > > Also, I'm almost as browser quirks fluent as this guy<http://www.quirksmode.org/>(though I'm not a very good writer, so no book to my credit), with one > exception: I'm better at XMLHttpRequest / iFrame / Script include wrangling > (ie: I found out good ways to fall back to other techniques if XMLHttp isn't > working and we're using JSON). So, I'd really like to have a stab at > debugging cross-browser compatibility issues. > > I'd also like to have a hand in separation of design / presentation / > structure / content / behavior of the OS (I've become pretty practiced in > it, so far). > > Oh, if anyone wants to see some stuff I've done recently: > http://wwwdev.wharton.upenn.edu/alumni (what I do at my real job - on the > development server 'cos it won't be live 'till monday. Take a look at the > JS for the navmenu and the tabbed boxes; they're great examples of how I do > JS OOP). > > http://www.fordi.org/comics (the neat little dock at the bottom is my > work; the pages are stuff I like to look at. For some reason, I'm not fond > of bookmarks. I'm actually working on it; a '+' symbol to add a link, and a > nice interface for choosing the image) > -- > Bryan Elliott > for...@gm... > http://www.fordi.org > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > _______________________________________________ > Atomos-discuss mailing list > Ato...@li... > https://lists.sourceforge.net/lists/listinfo/atomos-discuss > > > |
|
From: Bryan E. <for...@gm...> - 2007-01-07 09:31:43
|
> Naturally we need JS hackers, but there's also some other critical > items to be looked at. You have some history with SLAX and were the > author of MyDSL Maker (Personally my single-most-admired web-based > tool). Danks ^_^. I have some ideas for a mimimalist environment within which AtomOS could work. S'why I was asking about the backend stuff before. > Would you be interested in taking on an AOS-specific > customized live CD? Would you like to hack on the AOS code? Sure, on both counts. I have a bit of experience hacking on things like the backend bash daemons in Slax and Ubuntu that handle disk recognition, process throttling, etc (could be better done in Perl or C, but hey; not everyone does perl, and writing it in a compiled lingo means distributing the source). Also, I'm almost as browser quirks fluent as this guy<http://www.quirksmode.org/>(though I'm not a very good writer, so no book to my credit), with one exception: I'm better at XMLHttpRequest / iFrame / Script include wrangling (ie: I found out good ways to fall back to other techniques if XMLHttp isn't working and we're using JSON). So, I'd really like to have a stab at debugging cross-browser compatibility issues. I'd also like to have a hand in separation of design / presentation / structure / content / behavior of the OS (I've become pretty practiced in it, so far). Oh, if anyone wants to see some stuff I've done recently: http://wwwdev.wharton.upenn.edu/alumni (what I do at my real job - on the development server 'cos it won't be live 'till monday. Take a look at the JS for the navmenu and the tabbed boxes; they're great examples of how I do JS OOP). http://www.fordi.org/comics (the neat little dock at the bottom is my work; the pages are stuff I like to look at. For some reason, I'm not fond of bookmarks. I'm actually working on it; a '+' symbol to add a link, and a nice interface for choosing the image) -- Bryan Elliott for...@gm... http://www.fordi.org |
|
From: Scott E. <ps...@gm...> - 2007-01-05 22:22:09
|
On 1/5/07, Weirdbro <wie...@gm...> wrote: > Hello, I'm Weirdbro. I mostly just mess around with Javascript, Java, PHP, > and Ruby, making an occasional Flash game. I'm interested in OOP theory, and > I ran a Linux Vmware Virtual Machine on my computer, and I like to imagine I > have a slight idea of how Linux and *nix systems are set up and how it runs > on a low level, because I read technology articles. > In truth, I've never made a successful program in Javascript, but with my > general OOP and programming knowledge, Scott likes what I've done. > > I'm 14, so I have a job, but it has nothing to do with technology, > unfortunately. Well, the last time I played hockey, and the last time my mother produced the team flag, I _was_ #14. 4 had been taken and 14 was the next best thing in my mind. So forget _14_ - it's all a matter of perspective. We rule. It's up to us to do so. If we don't... who will? -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Weirdbro <wie...@gm...> - 2007-01-05 22:15:30
|
Hello, I'm Weirdbro. I mostly just mess around with Javascript, Java, PHP, and Ruby, making an occasional Flash game. I'm interested in OOP theory, and I ran a Linux Vmware Virtual Machine on my computer, and I like to imagine I have a slight idea of how Linux and *nix systems are set up and how it runs on a low level, because I read technology articles. In truth, I've never made a successful program in Javascript, but with my general OOP and programming knowledge, Scott likes what I've done. I'm 14, so I have a job, but it has nothing to do with technology, unfortunately. Weirdbro On 1/5/07, Scott Elcomb <ps...@gm...> wrote: > > On 1/4/07, Bryan Elliott <for...@gm...> wrote: > [...] > > So, yeah. Greetings. Howdy. 'Sup. Someone give me a project, > seriously. > > ^_^ > > A project indeed! > > Weirdbro and Digimer will hopefully be along with messages over the > few days - but in the meantime I do have some project-type stuff I'd > like to set on your plate. I just can't do that quite yet... Like > all of us, you have a diverse background - and one that I'd like to > leverage at that. :P > > As far as Atomic OS goes, everyone here is more-or-less an > All-Hands-Person. Weirdbro is mostly helping me with JavaScript > hacking, and Digimer is interested in AOS as a front-end UI solution. > I'll be hacking on their input for the v0.1.0 release, which I intend > to release at the end of this month. > > Is there anything in particular about Atomic OS that "turn's your > crank?" By which I mean, Is there any one or two things in specific > that you'd like to work on? > > Naturally we need JS hackers, but there's also some other critical > items to be looked at. You have some history with SLAX and were the > author of MyDSL Maker (Personally my single-most-admired web-based > tool). Would you be interested in taking on an AOS-specific > customized live CD? Would you like to hack on the AOS code? > > I won't ask for anyone to work on anything they don't want to or don't > feel comfortable with. To avoid that situation I'd like to know what > part(s) of AOS you would like to work on. ;-) > > > To all: Things have been really crazy the last 8 months - everything > from getting really, really sick to the loss of my brother to a > decently publicized FOSS project (_this_ project). I'll be away the > next few weekends, but momentum seems to be picking up. My Google > Alerts are popping up new posts, comments, and opinions about Atomic > OS on a fairly regular basis. I'd like to organize this like a > professional project -- even if we are doing it "part-time." ;-) > > Take care, and see you all in a few days! > - Scott. > > -- > Scott Elcomb > http://atomos.sourceforge.net/ > http://search.cpan.org/~selcomb/SAL-3.03/ > http://psema4.googlepages.com/ > > "They that can give up essential liberty to obtain a little temporary > safety deserve neither liberty nor safety." > > - Benjamin Franklin > > '"A lie can travel halfway around the world while the truth is putting > on its shoes." > > - Mark Twain > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys - and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Atomos-discuss mailing list > Ato...@li... > https://lists.sourceforge.net/lists/listinfo/atomos-discuss > |
|
From: Scott E. <ps...@gm...> - 2007-01-05 20:04:54
|
On 1/4/07, Bryan Elliott <for...@gm...> wrote: [...] > So, yeah. Greetings. Howdy. 'Sup. Someone give me a project, seriously. > ^_^ A project indeed! Weirdbro and Digimer will hopefully be along with messages over the few days - but in the meantime I do have some project-type stuff I'd like to set on your plate. I just can't do that quite yet... Like all of us, you have a diverse background - and one that I'd like to leverage at that. :P As far as Atomic OS goes, everyone here is more-or-less an All-Hands-Person. Weirdbro is mostly helping me with JavaScript hacking, and Digimer is interested in AOS as a front-end UI solution. I'll be hacking on their input for the v0.1.0 release, which I intend to release at the end of this month. Is there anything in particular about Atomic OS that "turn's your crank?" By which I mean, Is there any one or two things in specific that you'd like to work on? Naturally we need JS hackers, but there's also some other critical items to be looked at. You have some history with SLAX and were the author of MyDSL Maker (Personally my single-most-admired web-based tool). Would you be interested in taking on an AOS-specific customized live CD? Would you like to hack on the AOS code? I won't ask for anyone to work on anything they don't want to or don't feel comfortable with. To avoid that situation I'd like to know what part(s) of AOS you would like to work on. ;-) To all: Things have been really crazy the last 8 months - everything from getting really, really sick to the loss of my brother to a decently publicized FOSS project (_this_ project). I'll be away the next few weekends, but momentum seems to be picking up. My Google Alerts are popping up new posts, comments, and opinions about Atomic OS on a fairly regular basis. I'd like to organize this like a professional project -- even if we are doing it "part-time." ;-) Take care, and see you all in a few days! - Scott. -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Bryan E. <for...@gm...> - 2007-01-04 22:06:19
|
Ok. Anyways, I'm Fordi (or Bryan, if you'll not have it ^_^). I'm essentially a tinkerer, and have been playing with Linux, HTML, Javascript, C/C++, and a number of other things for what seems like aeons now. I wrote the web-based MyDSL maker (PHP), that is now defunct for a number of reasons (mostly that ibiblio got a little annoyed at me), an open source SRAM patcher for Gameboy Advance games (ANSI C), and a nigh-finished (it's a continuing background project) lib for editing ISOs in-place (ANSI C). While I was tinkering with Slax last year, I contributed a series of custom-built modules for Slax 5.1.6 called 'Rockstar', whose purpose was to patch Slax 5.1.6 for professional audio work, and provide the software with which it could be done. This involved a number of kernel patches, performance tweaks, ALSA/JACK wrangling, and from-source-only i686 compilation of every lib and program used. Slax Rockstar Edition (5.1.6) is still what I use for recording music. I play guitar, read corny webcomics and sci-fi novels, play the odd flash game, and am presently under the control of Final Fantasy 12 (which my brother got for me for christmas). Otherwise, I do all that other stuff that one's supposed to do in one's free time (eat, sleep, work out, shower, go to job, rinse, repeat). The job is a split one. I spent five years part-time in the Business and Public Policy department of the Wharton School as an administrative assistant. Recently, Wharton's External Affairs department has hired me, also part time, as a web developer. In sum, I work full time hours, but am still lacking the free benefits of actual full time status. So, yeah. Greetings. Howdy. 'Sup. Someone give me a project, seriously. ^_^ |
|
From: Scott E. <ps...@gm...> - 2007-01-04 03:44:51
|
Hi all, First off, I'd like to send a belated Happy Holidays to the list. Hope you all had a good Holiday Season. Next, I'd like to introduce a new developer to the project - Fordi - who has some Linux and Web Development experience. Finally, I'd like to ask all developers on the project to introduce themselves and their backgrounds in this thread. Primarily to give all of us an idea of our backgrounds so that we're all aware of what our combined skills look like. As the project manager I guess I should start off, so... I'm Scott, and Atomic OS is my poison-of-choice. Over the last 20 years or so, I've developed a number of applications and utilities in various languages (from C/C++ to Pascal/Delphi/Kylix, to Basic/VB/RealBasic, to Assembly, to AJAX, to...), one real-mode x86 OS called GEOS (General Expirements in Operating Systems - not the GEOS returned as a search result by Google or other search sites), and a game engine called SAGE (Simple Adventure Game Engine) done in DarkBasic for Windows. Over the last 4 or 5 years, I've focused on learning the FOSS (Free & Open Source Software) development paradigm and methodologies. Atomic OS is the result of that research and the need to make web-based user interfaces more intuitive - based on the desktop and command-line UI paradigms. Particular interests are: Furthering the spread of fundamental Computer Science knowledge and FOSS development strategies in general, as well as Social Science and OS/Web/Game development techniques in specific. -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Scott E. <ps...@gm...> - 2006-12-20 10:31:07
|
If you haven't been by the Atomic OS homepage in the last few weeks, you might wish to take a look. It's not perfect by any means, but it's a bit better than it was. Also, attached is a tarball I put together over the last 8 hours or so. It's a really, _really_ cutback edition, based on release v0.0.2 but including the OOP work that Weirdbro's put in. There is no console, shell, desktop, or anything else fancy. Just a bunch of tests and simple build structure. It also features a pure DHTML build tool, so you don't even need to have Perl installed to build your own versions. When it's ready for release, I'll make sure there are at least 2 release packages - one for *nix/*bsd and another for Windows. Comments would be appreciated as always. -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin '"A lie can travel halfway around the world while the truth is putting on its shoes." - Mark Twain |
|
From: Weirdbro <wie...@gm...> - 2006-10-14 22:56:21
|
Well, I'd like to suggest that we add a complete rewrite of DOMFS as a goal for version 0.1.0. If we move the window manager out, it needs to get rewritten, anyway. On 10/14/06, Scott Elcomb <ps...@gm...> wrote: > > Hi all, > > First off, I'd like to apologize for not being very active the last > few weeks. Between the Gmail/SourceForge mailing list gaff, and a > very hectic schedule, I just haven't been able to apply myself as much > as I'd hoped. > > Anyway, here's an early draft for a development roadmap. I'd like to > start incorporating Weirdbro's efforts into the main development tree. > > With that in mind, any comments on the following? We can assign some > dates down the road. Just curious if all the major points are covered > for now. > > [-- draft starts --] > Atomic OS Development Roadmap > > v0.1.1 - Release date TBD > - Process Management > - Device Files (Streams) > > v0.1.0 - Release date TBD > - Convert procedural design to OOP > - Remove non-critical items from the kernel code (eg. Window > Manager) > - Apply Terminal code > > v0.0.2 - July 21, 2006 > - Remove un-necessary media files, reduce release package size > / bandwidth usage. > > v0.0.1 - July 12, 2006 > - Initial Release > - Reverse engineered WAJAX to split single source file into > multiple files, improve coding standard > [-- draft ends --] > > -- > Scott Elcomb > http://atomos.sourceforge.net/ > http://search.cpan.org/~selcomb/SAL-3.03/ > http://psema4.googlepages.com/ > > "They that can give up essential liberty to obtain a little temporary > safety deserve neither liberty nor safety." > > - Benjamin Franklin > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job > easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > _______________________________________________ > Atomos-discuss mailing list > Ato...@li... > https://lists.sourceforge.net/lists/listinfo/atomos-discuss > |
|
From: Scott E. <ps...@gm...> - 2006-10-14 19:25:10
|
Hi all,
First off, I'd like to apologize for not being very active the last
few weeks. Between the Gmail/SourceForge mailing list gaff, and a
very hectic schedule, I just haven't been able to apply myself as much
as I'd hoped.
Anyway, here's an early draft for a development roadmap. I'd like to
start incorporating Weirdbro's efforts into the main development tree.
With that in mind, any comments on the following? We can assign some
dates down the road. Just curious if all the major points are covered
for now.
[-- draft starts --]
Atomic OS Development Roadmap
v0.1.1 - Release date TBD
- Process Management
- Device Files (Streams)
v0.1.0 - Release date TBD
- Convert procedural design to OOP
- Remove non-critical items from the kernel code (eg. Window Manager)
- Apply Terminal code
v0.0.2 - July 21, 2006
- Remove un-necessary media files, reduce release package size
/ bandwidth usage.
v0.0.1 - July 12, 2006
- Initial Release
- Reverse engineered WAJAX to split single source file into
multiple files, improve coding standard
[-- draft ends --]
--
Scott Elcomb
http://atomos.sourceforge.net/
http://search.cpan.org/~selcomb/SAL-3.03/
http://psema4.googlepages.com/
"They that can give up essential liberty to obtain a little temporary
safety deserve neither liberty nor safety."
- Benjamin Franklin
|
|
From: Weirdbro <wie...@gm...> - 2006-10-07 01:16:50
|
Its at weirdbro.com/src3.tar.gz It uses no config files yet. It works. |
|
From: Scott E. <ps...@gm...> - 2006-10-05 20:20:51
|
This is a test. According to the ticket, the issue between Gmail and SourceForge mailing lists has been resolved. Anybody out there? =) - Scott. ---------- Forwarded message ---------- From: SourceForge.net <no...@so...> Date: Oct 5, 2006 12:43 PM Subject: [ alexandria-Support Requests-1567803 ] Mailing list bounces (atomos-discuss) To: no...@so... Support Requests item #1567803, was opened at 2006-09-29 11:13 Message generated for change (Comment added) made by fincher You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=200001&aid=1567803&group_id=1 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: Project Mailing Lists/Archives/Services Group: Second Level Support >Status: Closed Priority: 5 Submitted By: Scott Elcomb (dleet) >Assigned to: Jeremy Fincher (fincher) Summary: Mailing list bounces (atomos-discuss) Initial Comment: I've had several messages bounced back in the last two days. Any help would be appreciated. =) [-- sample begins --] This is an automatically generated Delivery Status Notification THIS IS A WARNING MESSAGE ONLY. YOU DO NOT NEED TO RESEND YOUR MESSAGE. Delivery to the following recipient has been delayed: ato...@li... Message will be retried for 1 more day(s) Technical details of temporary failure: TEMP_FAILURE: SMTP Error (state 9): 451-Could not complete sender verify callout 451-Could not complete sender verify callout for <ps...@gm...>. 451-The mail server(s) for the domain may be temporarily unreachable, or 451-they may be permanently unreachable from this server. In the latter case, 451-you need to change the address or create an MX record for its domain 451-if it is supposed to be generally accessible from the Internet. 451 Talk to your mail administrator for details. [-- sample ends --] ---------------------------------------------------------------------- Comment By: Jeremy Fincher (fincher) Date: 2006-10-05 12:43 Message: Logged In: YES user_id=1019020 Greetings, To the best of my knowledge, this matter has now been resolved. Should you require further assistance from the SourceForge.net team, please add a comment to this Support Request and we will reopen it for you. SourceForge.net Support ---------------------------------------------------------------------- Comment By: Billy B (baxissimo) Date: 2006-10-02 21:46 Message: Logged In: YES user_id=821871 The following issues in your bug tracker ALL concern this problem: http://sourceforge.net/tracker/index.php?func=detail&aid=1569441&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1569366&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1569326&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1569138&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1569055&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568827&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568600&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568577&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568565&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568458&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568306&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568223&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568118&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1568009&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1567854&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1567803&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1567745&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1567366&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1567120&group_id=1&atid=200001 http://sourceforge.net/tracker/index.php?func=detail&aid=1566582&group_id=1&atid=200001 ---------------------------------------------------------------------- Comment By: Scott Elcomb (dleet) Date: 2006-09-29 19:51 Message: Logged In: YES user_id=739050 wdavison wrote: > I assume your email address is in the gmail.com domain, > correct? Google apparently made a change that is causing > problems for a lot of people who run the exim mail > software. We're looking for a solution. Yes - and fortunately - I only have one other active project member (also gmail) to worry about at the moment. I can imagine other project admins are going crazy though. Thanks for supporting us all. =) ---------------------------------------------------------------------- Comment By: Wayne Davison (wdavison) Date: 2006-09-29 17:39 Message: Logged In: YES user_id=1546419 Greetings, I assume your email address is in the gmail.com domain, correct? Google apparently made a change that is causing problems for a lot of people who run the exim mail software. We're looking for a solution. SourceForge.net Support ---------------------------------------------------------------------- Comment By: Luca Dionisi (lukisi71) Date: 2006-09-29 11:32 Message: Logged In: YES user_id=1590314 I'm another one from gmail. Same here. ---------------------------------------------------------------------- Comment By: Dave Hylands (dhylands) Date: 2006-09-29 11:21 Message: Logged In: YES user_id=262293 I'm seeing a similar problem, also from a gmail account. Non-gmail accounts seem to be working fine ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=200001&aid=1567803&group_id=1 -- Scott Elcomb http://atomos.sourceforge.net/ http://search.cpan.org/~selcomb/SAL-3.03/ http://psema4.googlepages.com/ "They that can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety." - Benjamin Franklin |
|
From: Weirdbro <wie...@gm...> - 2006-09-27 02:14:51
|
Which executable commands don't work? I was sure I had fixed most of them, the only one I don't have is `cd`, since it isn't an executable, but a real WASH command. If you've noticed, I've implemented a environment variable system where processes, if they use the this.exec command, copy their env variable sinto their child. I didn't really think about a root process. How do you think it should work, why is it necessary to be able to access it, and what would other processes use that for? Processes, because of the inability in our current system to safely remove events, need to have a function that is called when it is killed. It needs to kill its own process when it finishes running, since we have no way of knowing whether it still has events attached, keeping it from being garbage collected by javascript. The build system, when it moves rom into files, it doesn't escape their contents, leaving '<' and '>' to be interpreted by Firefox as elements, which absolutely messes up their innerHTML, because the stuff between < and > is interpreted by firefox as element attributes, and has ="" added after every space. That was incredibly hard to figure out, and I just added about 2 lines of code to the ROM builder to make sure it escapes stuff before it writes the file. Also, I had to write some DOMFS functions to handle the same things for reading and writing DOMFS files from Javascript. Have you thought about reimplementing hte build system in Javascript? It fits with Atomic OSs philosophy, is much more cross-platform than a Perl and BASH setup, and it just cleaner, at least if we reimplement it cleaner. We really need to write the IO objects and add tons of functions to DOMFS, so that even if it is based off of widgets, they should never have to use widget functions ever outside of writing kernel code. -Weirdbro On 9/26/06, Scott Elcomb <ps...@gm...> wrote: > > On 9/25/06, Weirdbro <wie...@gm...> wrote: > > Here's my latest version: > > http://www.weirdbro.com/src2.tar.gz > > > > I didn't finish adding objects to the kernel, but I added a Process > class, > > moved Wash out of the kernel, fixed a glitch in the build system, > > and started moving the commands from wash out to binaries. `cd` needs to > be > > readded to Wash. > > Processes are a huge step forward. *Thank you* > > Most of the executable commands no longer work, but that's the way it > is. We also need garbage collection in the process object. When > child processes complete their code, they need to be "killed" so we > don't end up using unnecessary resources in the client browser. > > Based on your code and style of programming, I think a Kernel object > is in order... Given that, I'd also suggest renaming the current > global Process object to Kernel.processZero or some other name > indicative of the Root Process found on *nix type systems. > > Uhmm, what glitch did you find in the build system? lol. I'm still > tied up looking at your code and trying to figure out the > implications... Thanks again! > > -- > Scott Elcomb > http://atomos.sourceforge.net/ > http://search.cpan.org/~selcomb/SAL-3.03/ > http://psema4.googlepages.com/ > > "They that can give up essential liberty to obtain a little temporary > safety deserve neither liberty nor safety." > > - Benjamin Franklin > > ------------------------------------------------------------------------- > Take Surveys. Earn Cash. Influence the Future of IT > Join SourceForge.net's Techsay panel and you'll get the chance to share > your > opinions on IT & business topics through brief surveys -- and earn cash > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > _______________________________________________ > Atomos-discuss mailing list > Ato...@li... > https://lists.sourceforge.net/lists/listinfo/atomos-discuss > |
