Squeak-dev January 2005

squeak-dev@lists.squeakfoundation.org

134 participants
235 discussions

Why I asked (was: How might I set the cursorpoint? )
by Peace Jerome 29 Jan '05

29 Jan '05

> >Peace Jerome <peace_the_dreamer at yahoo.com> wrote: > >> Tim wrote: >> >> >> >Prim 91 is actually test display depth... we don't >> do setting of the cursor position. >> >> That's interesting. > >It's very old and ought to be removed so as to avoid confusing people. I went back and found where Prim 91 is used for test display depth. So you gave me the answer I was looking for. Setting cursorpoint is off the menu for everyone until its reinvented. > > >> >It'sa pretty bad bit of UI generally to take the >> users >> >'attention' and drag it somewhere else by force. >> > >> >> I pretty much agree with you. To see what I was trying >> to do. Get a rectangle ask for the bitmap fill style >> and then ask (via red menu) to reset the origin (or >> orientation). The cursor is where the menu item was >> selected which is usually nowhere near the original >> origin. So the UI is that the picture changes >> dramatically to jump the origin to where the cursor >> now is. It is not to bad for just the origin but the >> orientation stuff is rather jarring. So I thought that >> something that leaves them where they are until you >> move the cursor would be more pleasant. I�ve been >> experimenting with different things that might do that >> and following up my curiosity.\ >Interesting problem. Seems to me that you should get a chance to >choose the new origin you want rather than it simply jumping to the >cursor. Yeah I�ve tried several things. One of them put the handle where the origin was. Then as soon as the cursor moved it snapped to the cursor. It seemed a little better than instant chaos but still not elegant. I�ve put the problem on the back burner til I figure out something else interesting to try. Maybe just give the adjustments their own handles and let the user play around then request their removal like with polygons. > >> >> I�ve found many a piece of code in squeak that >> indicated someone intended for a particular function >> to work. > >Ah, you've spotted the real weakspot of Squeak - too much left over >crud from unfinished and abandoned experiments and no documentation of >intent. Infuriating isn't it? Now this is the part of the note I wanted to respond to. The thought that someone worked on squeak an then abandon their work and squeak has been on several occasions infuriating. The broken code itself isn�t infuriating. Its actually rather informative of intent. Play with morphic. Sense a feature is missing. Look at the code. Find an if statement hiding the feature. The code the if statement hides is usually the feature broken but included. Now its just a debugging problem. And thats solvalbe with enough patience and time. It would be better if squeak were full of consistently good quality code. To show by example what should be done. The poetry of how the core squeak classes work is sort of neat and inspirational. The problem code is challenging, frustrating yet also educational and when I can do something about it, a rewarding sensation of accomplishment. What�s infuriating are the people problems. How to get significant contributions published into the squeak stream. How to get them acknowledged as good and needed contributions. How to get them acted upon. I want the stuff I improve to be available for others to use and improve. The feeling that the caretakers don�t really feel the need to take really good care of squeak and that there is no good way of shaking them until they do. Now that�s infuriating. > >> And while I�m on the subject of curiosity who are the >> we of >> >> > ... we don't do setting of the cursor position. >> >> Who makes the communities standards? What are they? >Muah-ha-ha! _I_ do! It's me! Bow before me oh feeble earthlings! > >Oh, and that Goran character. He pulls the strings via a complex web of >blackmail and coercion, drugs and surveilance. All under the control of >the Evil Overlord of course. > Ah yes looney toons. They are all looney toons. Another question answered. > Cheers and Joy, --Jerome Peace __________________________________ Do you Yahoo!? All your favorites on one personal page � Try My Yahoo! http://my.yahoo.com

1 0

RE: [Maybe Spam] Re: Swiki vandalized
by Lex Spoon 28 Jan '05

28 Jan '05

"Frank Shearar" <Frank.Shearar(a)rnid.org.uk> wrote: > That doesn't follow. Say you're very skilled with proofreading and editing, with only a cursory knowledge of Smalltalk. Thus, you're eminently qualified to WikiGnome, but don't have a grasp of the community coding standards, and thus shouldn't (yet) be allowed to push changesets to the update stream. > > _I_ sure as heck am not conversant enough with Squeak to try push changesets to the update stream (in general) even if I could. In fact, quite a lot of the intended usage of the wiki is for complete novices who would have a low SqueakPeople ranking. For example: 1. Newbie pages are, duh, for newbies. 2. When someone asks a question, we often ask that they go summarize their answer on the wiki. Again, these people are often newbies. So, I dislike basing editing privilage on your programming prowess. Other dislikes: There are a lot of schemes that are tempting due to security by obscurity -- for example, requiring that people have registered on SqueakPeople is already going to shut down gazillions of bots that just won't bother -- but security by obscurity bugs me. Additionally, there are a lot of schemes that stop stupid bots, e.g. requiring someone to decode a GIF, but those bug me even worse because they just seem to breed ever better bots. I have a proposal in the next message. -Lex

2 2

Let us face reality
by stéphane ducasse 28 Jan '05

28 Jan '05

Hi all There is no point to send fix or enh in the mailing-list if no one (or only bert) is looking at them and collecting them. Marcus got burnt (really burnt to the point to be really incapable to think about his research) in the past. I helped and I will continue as soon I can and 3.8 is out. But if we do not find a way to identify resources (= pay someone to do something) so that this job is done, then forget Squeak. We are trying to get a real foundation on foot but I do not see a real interest. Let us go and hack our own fork this is more fun.... Stef

10 17

RE: Let us face reality
by Frank Shearar 28 Jan '05

28 Jan '05

stéphane ducasse <ducasse(a)iam.unibe.ch> wrote: > > Hi all > > There is no point to send fix or enh in the mailing-list if > no one (or > only bert) is looking at them and collecting them. Marcus got burnt > (really burnt to the point to be really incapable to think about his > research) in the past. I helped and I will continue as soon I can and > 3.8 is out. But if we do not find a way to identify resources (= pay > someone to do something) so that this job is done, then > forget Squeak. > We are trying to get a real foundation on foot but I do not > see a real > interest. > > Let us go and hack our own fork this is more fun.... We actually need a brief on what it means to be a Good Citizen(tm) in this community. Not just being polite and all that social stuff, but what the community needs from us as members. Reviews? Fixes? Bug reporting? So. Stephane, does your frustration stem from the lack of reviewing that happens? (I'm sure you've answered this before many times, but this topic's come up so many times these past few months that it's difficult to see the wood for the trees.) Some random thoughts: * We need people to find bugs. * Then we need people to write tests for those bugs. * Then we need people to write fixes for those tests. We don't need reviews of these fixes - if they pass the tests, then they fix the problem! * Then we need people to commit those fixes (and tests) to the update stream. On a slightly different aspect, * We need people to provide cool new features for inclusion in the base image. * Then we need people to review these new features. * Then we need people to commit these enhancements to the update stream. Finally, * We need people to write tests for the base image. * Then we need people to use those tests to write refactorings. * Then we need people to check the refactorings, making sure they don't break packages (or if they do, to mail the affected maintainers). * Then we need people to commit these refactorings to the update stream. Out of the above, do you assert that we need more reviewers? frank ******************************************************************* This email and any files transmitted with it are confidential and intended solely for the use of the individual or entity to whom they are addressed. Any views or opinions expressed are solely those of the author and do not necessarily represent RNID policy. If you are not the intended recipient you are advised that any use, dissemination, forwarding, printing or copying of this email is strictly prohibited. If you have received this email in error please notify the RNID Helpdesk by telephone on: +44 (0) 207 296 8282. The Royal National Institute for Deaf People Registered Office 19*23 Featherstone Street London EC1Y 8SL No. 454169 (England) Registered Charity No. 207720 ********************************************************************

4 4

Harvesting Party Tuesday February 1st 08:00 GMT
by Ken Causey 28 Jan '05

28 Jan '05

Just a reminder that the next Harvesting Party will take place between 08:00 (8 AM) and 12:00 (Noon) GMT Tuesday February 1st, 2005. As usual we will meet in #squeak on irc.freenode.net. More information about the #squeak IRC channel can be found at http://people.squeakfoundation.org/article/7.html Information about the Harvesting Party idea can be found at http://people.squeakfoundation.org/article/29.html If you have the time please attend and work toward the future of Squeak. The next seven parties are scheduled for 12:00 (12:00 PM) GMT Saturday February 05, 2005 16:00 (04:00 PM) GMT Wednesday February 09, 2005 20:00 (08:00 PM) GMT Sunday February 13, 2005 00:00 (12:00 AM) GMT Friday February 18, 2005 04:00 (04:00 AM) GMT Tuesday February 22, 2005 08:00 (08:00 AM) GMT Saturday February 26, 2005 12:00 (12:00 PM) GMT Wednesday March 02, 2005

1 0

Re: majorShrink
by Lex Spoon 28 Jan '05

28 Jan '05

Kamil Kukura <kamk(a)volny.cz> wrote: > I was pondering about #majorShrink as about some tool for image > customization. That it would have interactive window asking what to > discard. Among other things there could be some nifties like: > > discardExperiments > > discardOldCrap > discardDeprecated > > Of course, it should be maintained with every version published which is > maybe hardest task... > You can do this right now, of course: just comment out the parts you wish to keep. :) There are certain kinds of tools you don't need, if you are immersed in a great programming environment to begin with. (The tool could do other things, yes, such as keep track of dependencies between the pieces.... But I doubt *anyone* is up for that much work. Far better to spend the time packagizing these components for real!) Lex

2 1

Help! Recovering/editing a buggy .image
by Daniel B. Faken 28 Jan '05

28 Jan '05

Hi all, I managed to save an image with a process running that causes a segfault: %squeak ./Croquet0.1.image tst.st Segmentation fault -1199407560 [] in SDLayer class>nextEvent: -1199407688 Integer>timesRepeat: -1199407780 SDLayer class>nextEvent: -1199407872 SDLayer class>checkEvents: -1200100432 [] in UndefinedObject>? -1200100340 [] in BlockContext>newProcess Abort (tst.st was my attempt to have a script run first) How can I 'edit' the image to remove this process (or at least change the nextEvent: method so it doesn't crash)? thanks, Daniel

2 1

[BUG]Array(Object)>>error:
by Raymond Asselin 28 Jan '05

28 Jan '05

opening SqueakMap package loader from the world menu I received an 'error occured updating map: primitive have failed' so I scrapped the sm directory and rebegin all the process...same error ....choose to not opening a debugger and SqueakMap window show in world. But trying to update map from the net the same error occured.... and seems to be related to array, referenceStream and/or ImageSegment...? It is really bad as this image is of real value for me... Any help is welcome. 26 January 2005 10:07:07 am VM: Mac OS - a SmalltalkImage Image: Squeak3.7 [latest update: #5989] SecurityManager state: Restricted: false FileAccess: true SocketAccess: true Working Dir IGod:Users:ray1:Documents:Bureau:Institut Universitaire Trusted Dir IGod:Users:ray1:Documents:Bureau:Institut Universitaire Untrusted Dir IGod:Users:ray1:Library:Preferences:Squeak:Internet:My Squeak Array(Object)>>error: Receiver: #(a SMSqueakMap) Arguments and temporary variables: aString: 'a primitive has failed' Receiver's instance variables: #(a SMSqueakMap) Array(Object)>>primitiveFailed Receiver: #(a SMSqueakMap) Arguments and temporary variables: Receiver's instance variables: #(a SMSqueakMap) Array>>elementsForwardIdentityTo: Receiver: #(a SMSqueakMap) Arguments and temporary variables: otherArray: #(a SMSqueakMap) Receiver's instance variables: #(a SMSqueakMap) ImageSegment>>reshapeClasses:refStream: Receiver: an ImageSegment Arguments and temporary variables: mapFakeClassesToReal: an IdentityDictionary(Fake37SMSqueakMap->SMSqueakMap ) smartRefStream: a SmartRefStream a ByteArray(39 70 114 111 109 32 83 113 117 10...etc... bads: an OrderedCollection(a SMSqueakMap) allVarMaps: an IdentityDictionary(a SMSqueakMap->a Dictionary(#ClassName->#SMSq...etc... perfect: a SMSqueakMap insts: #(a SMSqueakMap) partials: an OrderedCollection(a SMSqueakMap) in: an OrderedCollection(a SMSqueakMap) out: an OrderedCollection(a SMSqueakMap) aFakeClass: Fake37SMSqueakMap theRealClass: SMSqueakMap obj: nil misShapen: a SMSqueakMap x: nil each: a SMSqueakMap Receiver's instance variables: arrayOfRoots: #(a SMSqueakMap) segment: a WordArrayForSegment(1929386342) outPointers: #(Fake37SMSqueakMap nil 'sm' true Dictionary SMFileCache Array an ...etc... state: #imported segmentName: nil fileName: nil endMarker: SmartRefStream>>next userRootCnt: 1 renamedClasses: nil --- The full stack --- Array(Object)>>error: Array(Object)>>primitiveFailed Array>>elementsForwardIdentityTo: ImageSegment>>reshapeClasses:refStream: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - ImageSegment>>comeFullyUpOnReload: SmartRefStream(DataStream)>>next SmartRefStream(ReferenceStream)>>next SmartRefStream>>next SmartRefStream>>scanFrom: ObjectScanner>>scanFrom: [] in RWBinaryOrTextStream(PositionableStream)>>fileInAnnouncing: {[val _ (self peekFor: $!) ifTrue: [(Compiler evaluate: self nextChunk lo...]} BlockContext>>on:do: [] in RWBinaryOrTextStream(PositionableStream)>>fileInAnnouncing: {[:bar | [self atEnd] whileFalse: [bar value: self position. self skipS...]} ProgressInitiationException>>defaultAction UndefinedObject>>handleSignal: MethodContext(ContextPart)>>handleSignal: MethodContext(ContextPart)>>handleSignal: ProgressInitiationException(Exception)>>signal ProgressInitiationException>>display:at:from:to:during: ProgressInitiationException class>>display:at:from:to:during: String>>displayProgressAt:from:to:during: RWBinaryOrTextStream(PositionableStream)>>fileInAnnouncing: RWBinaryOrTextStream(PositionableStream)>>fileIn RWBinaryOrTextStream(ReadWriteStream)>>fileInObjectAndCode [] in SMSqueakMap>>reload {[map _ stream fileInObjectAndCode install arrayOfRoots first]} BlockContext>>ensure: SMSqueakMap>>reload SMSqueakMap>>synchWithDisk SMSqueakMap>>loadUpdatesFull: SMSqueakMap>>loadFull SMSqueakMap>>loadUpdates [] in SMLoader>>loadUpdates {[squeakMap loadUpdates. self noteChanged]} BlockContext>>ensure: CursorWithMask(Cursor)>>showWhile: [] in SMLoader>>loadUpdates {[Cursor wait showWhile: [squeakMap loadUpdates. self noteChanged]]} BlockContext>>on:do: SMLoader>>loadUpdates SMLoader>>perform:orSendTo: [] in MenuItemMorph>>invokeWithEvent: {[(selArgCount _ selector numArgs) = 0 ifTrue: [target perform: selector] ...]} BlockContext>>ensure: CursorWithMask(Cursor)>>showWhile: MenuItemMorph>>invokeWithEvent: MenuItemMorph>>mouseUp: MenuItemMorph>>handleMouseUp: MouseButtonEvent>>sentTo: MenuItemMorph(Morph)>>handleEvent: MorphicEventDispatcher>>dispatchDefault:with: MorphicEventDispatcher>>dispatchEvent:with: MenuItemMorph(Morph)>>processEvent:using: MorphicEventDispatcher>>dispatchDefault:with: MorphicEventDispatcher>>dispatchEvent:with: MenuMorph(Morph)>>processEvent:using: MenuMorph(Morph)>>processEvent: MenuMorph>>handleFocusEvent: ....etc...

3 11

Re: [Maybe Spam] Re: Swiki vandalized
by Lex Spoon 28 Jan '05

28 Jan '05

Here's an approach I haven't seen described yet: have a wiki-editing rank, but base your privilage of editing on your history of making acceptible edits. That is, you increase in rank by making good edits, and you decrease (dramatically) in rank by spamming. Has anyone heard of such a system? Here's a straw man system just to give you the idea. There is a lot of tweaking that is possible. There are three ranks: RandomSchmoe. Anyone off the net. You do not even need to log in. Editor. Someone who is allowed to edit pages, due to a history of acceptible edits. Moderator. Someone who has been long been a good citizen, and can start allowing others into the community or kicking others out. As a Random Schmoe, you can view all you want. If you edit a page, then your proposed edit goes into a queue for moderation. Whenever you queue a proposed edit, you can create an account if you want; or, you can leave it as an anonymous edit if you prefer. (This latter is good for stupid spelling fixes, or small corrections by outside people who really have no plan to return to our wiki. I want to avoid red tape as much as possible!) If a Moderator approves 1 proposed edit of yours, then you become an Editor. If you log in and you are an editor, then you can edit all the pages you like. If you make at least 10 edits, and you are in the community at least 1 month, then you automatically become a Moderator. If you make an unacceptible edit, then 2 moderators may collude to remove your editing privilages. You can also approve queued edits that schmoes have made, but the schmoe will remain a schmoe. If you are a Moderator, then you are a keeper of the gates of the community. You can elevate a schmoe to an editor by pointing to any proposed edit that you approve of. If an editor beat you to it, you can re-approve (including one that an editor has already approved This is a basic scheme. To fancy it up, there could be multiple levels, each with a *rate* of editing that is allowed. So editor-10 can make 10 edits per day, and editor-100 can make 100. You have to make 100 total edits before attaining editor-100. Something like that. Also, rank should probably degrade over time if you don't do anything on the wiki. I want the ranks to represent the *active* community. Also, size of edits should matter; I'm just talking about number of edits, to simplify the discussion. I haven't said exactly how proposed edits are *rejected* when they are bad. We need a mechanism for that, and there are many ways to do it. You might simply leave things on a queue for 5 days. You might let moderators whack edits immediately, or two editors to get rid of one. Or you might let anyone get rid of a proposed edit, but the edit gets logged anyway. I hope the specific approach doesn't matter: there won't be many bots attacking a wiki of this kind. The general principles of this approach are: 1. Spambots don't work. 2. To abuse the system, you have to precede the abuse by an approximately equal amount of good deeds. 3. Goodness vs. abuse is decided by active members of the community. 4. The red tape stays out of the way for two important operations: viewing, and tiny fixes. 5. Moderators are mostly used to flag *bad* edits, not good ones, because good edits are far more common. (With the notable exception of edits by complete schmoes, where a moderator must flag the edit as good.) The main drawback is that there needs to be a little bit of moderation to raise people to Editor. Given our problems with BFAV non-participation, this might be a problem. However, wiki moderation seems likely to be very easy compared to BFAV bug evaluation. Further, I hope that most edits are done by repeat editors; moderators only get involved in the first edit that people do, and thus I hope not a lot of moderation is needed. -Lex

1 0

Another really cheap XScale Linux SBC for Squeak
by Jon Hylands 28 Jan '05

28 Jan '05

Hi everyone, On one of the robotics groups I'm on, someone brought up the following consumer device: Linksys NSLU2 http://www.linksys.com/products/product.asp?grid=35&scid=43&prid=640 Its apparently got a 200 MHz XScale inside, 32 MB of RAM, with a couple USB host ports and an ethernet plug, and it runs Linux natively inside, and it is quite hackable. It should probably take very little effort to get Squeak running on this machine. Here are some more links: http://www.batbox.org/nslu2-linux.html http://groups.yahoo.com/group/nslu2-linux http://www.nslu2-linux.org http://www.tomsnetworking.com/Sections-article85.php Of course, the really cool thing about this machine is the price -- you can get them new from http://www.buy.com for $80 USD. Later, Jon -------------------------------------------------------------- Jon Hylands Jon(a)huv.com http://www.huv.com/jon Project: Micro Seeker (Micro Autonomous Underwater Vehicle) http://www.huv.com

2 1

← Newer
1
2
3
4
5
6
7
8
...
24
Older →

Jump to page: