From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me X-Spam-Level: X-Spam-Status: No, score=-1.9 required=5.0 tests=BAYES_00 autolearn=ham autolearn_force=no version=3.4.4 X-Google-Language: ENGLISH,ASCII X-Google-Thread: 103376,1943b1e68472411f X-Google-Attributes: gid103376,public X-Google-ArrivalTime: 2002-06-12 15:34:30 PST Message-ID: <3D07CA18.EB2CAF55@sympatico.ca> From: David Marceau X-Mailer: Mozilla 4.79 [en] (X11; U; Linux 2.4.17-10mdksmp i686) X-Accept-Language: en MIME-Version: 1.0 Newsgroups: comp.lang.ada Subject: Re: Why write an Ada web browser ?, was: Re: GNAT Ada - DLL - MSVC References: <3D062F7D.406B8709@sympatico.ca> <3D0761FF.38554417@sympatico.ca> <3D076F21.E0CB708E@san.rr.com> <3D0798DE.9F54B0D0@sympatico.ca> <3D07A181.4E478A0E@san.rr.com> Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Date: Wed, 12 Jun 2002 18:24:24 -0400 NNTP-Posting-Host: 65.92.168.142 X-Complaints-To: abuse@sympatico.ca X-Trace: news20.bellglobal.com 1023920828 65.92.168.142 (Wed, 12 Jun 2002 18:27:08 EDT) NNTP-Posting-Date: Wed, 12 Jun 2002 18:27:08 EDT Organization: Bell Sympatico Path: archiver1.google.com!news1.google.com!newsfeed.stanford.edu!news.ems.psu.edu!news.cis.ohio-state.edu!news.maxwell.syr.edu!sunqbc.risq.qc.ca!torn!webster!nf1.bellglobal.com!nf2.bellglobal.com!news20.bellglobal.com.POSTED!not-for-mail Xref: archiver1.google.com comp.lang.ada:25834 Date: 2002-06-12T18:24:24-04:00 List-Id: > I think we're talking at cross purposes. Unless you're going to reimplement > *every* plug-in in Ada, the plug-ins you don't reimplement will have the > same security problems. That is exactly the intent. Change the architecture of the main application the web browser, and change the architecture of the subsystems the Flashplayer and whatever else plugins matter to you. This implied re-writing the web browser in ada. This implied re-writing the flash player in ada. The way the system would work. 1)user invokes the web browser 2)user invokes some flash file by opening it. 3)the web browser detects the type for the file to open 4)the web browser then invokes into another process a decoration-less/decorated window placed within/outside the web browser window. 5)the flash plug-in loads the desired user file via the protocol between the two processes. Note how the plug-in accesses its resources. Via the protocol back to the web browser. Note the web browser commands the file the plug-in is to use. Note the plug-in asks for URL'S from the web browser. it can't open them itself. Anything else the plug-in tries to do forget it. To draw, the plug-in asks the web browser api. The window device context never directly gets into the hands of a plug-in. To get input/output from other devices(i.e. mouse), the plug-in would ask the web browser api. The device itself never gets into the hands of the plug-in. In other words the plug-in would request a drawing/filewriting/filereading/mousereading service from the web browser. The web browser with its multiple tasks would then actually do the drawing/filewriting/filereading/mousereading after doing some security checks. The web browser owns the mouse and whatever else devices. The plug-in owns nothing. Doesn't matter whether it's a flash player or a video player ... > > That is, sure, you can implement a Flash player or a ReadVideo player in Ada > and perhaps fix some security bugs by doing so. But making your web browser > secure and still having it invoke RealVideo's player isn't going to be any > safer. I beg to differ since I have from the beginning implied re-writing everything browser and all plug-ins. IMHO It would be safer more maintainable and more platform-independant reusable. Note how I didn't say faster. One more thing someone implied re-writing a jvm in ada. I will clarify this. The intent is not to run java-byte-code in the so-called ada-web-browser. The intent is to purely run ada code in the ada-web-browser and preferrably straight from source one builds himself. Yes the plug-in mechanism should provide something for downloading all the source for a new plug-in and recompile on the machine. Yes this implies gnat-ada-web-browser-plug-in-compiler which is I'm guessing something that would verify that the plug-in source code only with's in calls from the ada-web-browser-plugin-api package and nothing else. Operating System specific calls would not get compiled in. How to do this? I don't know. Maybe something like j2me(java language subset for cell phones and palm pilots) code converter/preverifier? It may be like a jvm but I don't think it is a jvm. The intention is to run the original and unique ada-run-time-environment which gets linked into the plug-in. Maybe the ada-rte could be considered an avm :) Maybe removing os-specific calls and adding the ada-web-browser-plugin-api to the ada-rte would more like an ADA Virtual MACHINE if there would be such a thing. I am just dreaming. When embedding stuff in html using why not provide url's to cvs source code server. Most of the unix world is doing this outside of their web browser for getting source all over the world anyways. I'm dreaming. I'm human. As you can see I'm venting my frustration with the status-quo so I apologize to anyone I have offended. How did I get to wanting a better browser? In windows internet explorer, I went to some web sites heavy with javascript i.e. www.act.com (the contact manager s/w) and it crapped out my windows os. I needed to reboot. On my linux box running netscape, I do the same thing. I go to www.act.com. It craps outs netscape, but not my linux box. I just xkill on netscape and just restart it up and everything's fine. At least in linux it's just annoying. In windows however it was intolerable. IMHO Seeing this kind of bug appear again and again after a couple of years, I believe this stuff shouldn't happen and the jvm/javascript engines along with the current web browser implementations are the source of these problems. That's how I got to dreaming of this Ada-Web-Browser and Ada-Web-Browser-Plugin idea :) Sant� bonheur, David Marceau