Literally every browser is unsafe, and has always been unsafe. There's zero reason this should be the case. Transferring documents online is not that hard a problem. Perhaps blinging out all the latest features for all the latest ad companies is not worth the cost to users' system security.
Even if every computer was running an operating system that was developed as carefully as equipment in charge of people's lives, I'm sure there'd still be vulnerabilities. The low hanging fruit is relative, and there'll always be people cracking computers.
The main problem is that what used to be called "mobile code" before smartphones were a thing is really convenient. That's why Java tried it too. Sandboxed code helps a lot, so there's this constant tension between trying to make the sandbox less restrictive and keeping it secure.
Sometimes I think that despite poor execution the JVM guys had the right idea. Sandbox code from the net, but also have code signing to fall back on.
As for rarely used, I'd guess Google Maps is a pretty well used site that uses WebGL.
You're right, Google Maps is a good example of WebGL use. But it could also just be a regular desktop app. People would download it just fine.
The web looks and works pretty much how it did 10 years ago. Almost every new feature is just a replacement for an old feature that did the same thing but not as well.
We've gone from plugins to add features to building the features right into the browser. We've gone from tens of megabytes to hundreds of megabytes just to render some text and images. There is literally too much code to audit. I think even kernel hackers would have trouble understanding the complexity of modern browsers.
In 20 years, do you expect the web will be significantly different than it is now? The interface might change, but you can't really innovate on text and pictures.
Although the length of your post worries me that you might actually believe what you are saying.
Please check out the Bananabread example on mozilla's developer page (just search it on your favorite search engine). Please do not post such comments again, there are people on here that will not realize you're joking.
> Please do not post such comments again, there are people on here that will not realize you're joking.
What is so dangerous about someone having the opinion that HTML shouldn't get more features? This "be careful what you say"-tone is so weird to me, on this topic.
> Bananabread
A 3D shooter in the browser. So what? Yes there are advantages to having this power, but not everyone is going to agree that it is worth it.
Having to support old tech lengthens development and testing time quite a bit. We need more browser features so we can write less code, only write it once, and distribute self-contained components.
A website is literally just text and pictures. And video. That's it. We really don't need it to be so massively complicated, to suck up so many resources, and to be such a security nightmare.
In 20 years, your browser will probably use a gigabyte or two of ram minimum, require four CPU cores, and consist of several hundred thousand lines of code - maybe a million.
To render text.
Yes and that is one of the exciting innovations being worked on right now, web components. That's exactly why I brought it up.
Some websites are just text and pictures, it is true. We call them brochure-ware. But not the sites I work on. I build applications for managing your entire business. They are just text in pictures in the same way that a PC is just text and pictures. Should we halt innovation on computers in general? NASA's flight control systems are really just calculators if you break it down as you have done with the web.
At the worst you sound anti-technology, and at best, snobbish about your area of specialty. Implying that whatever you work on is important, but things others work on are 'just text'.
Aah, a Luddite, seize him!
Maybe it's not anti-technology as much as anti-cram-everything-into-the-browser. Some people think that full-fledged applications should be kept out of the browser, but that doesn't mean that they think that full-fledged applications shouldn't exist at all.
And CSS, which supports transform and animation and is inching towards Turing completeness, and javascript, and maybe HTML5/Canvas and WebGL.
Like as not, we seem to be converging towards a point where websites are as much applications as documents.
I'm not saying this excuses the industry, but that doesn't change incentives. If you want more secure services, you need to change the preferences of the users.
throw it all away and rewrite it all again in two years when even newer features get added to the browsers!
Here's an idea: make your own browser that competes with Chrome for usability, offer a huge prize for anyone that hacks it. It's easy, right?
Transferring is the easy part. Rendering them is the hard part.
Suddenly it doesn't sound so reasonable.
CP67 was an abbreviation for control program 67 for the IBM 360/67 computer of, right, about 1967 and was developed by the IBM Cambridge Scientific Center as a means of interactive, time-sharing development of operating systems.
Later commercial time-sharing services used CP67. So, could have a few dozen users, each writing whatever code they wanted, e.g., assembler since it was a good environment to make developing assembler code easy, with, as far as I know, no user ever hurting the CP67 code or the work of any other user.
So, right, CP67 was the first or one of the first cases of virtual machine. Its security provisions appeared to be absolute -- run any code at all, including privileged code, e.g., including operating systems, including CP67 itself, with full safety and security.
Two years or so later there was Multics from MIT Project MAC. It featured capabilities and attribute control lists (still with us). As far as I know, they worked fine. For some years there was a Multics in the basement of the Pentagon and regarded as secure computing. Later Prime Computer did something quite similar and claimed that they had a prize for anyone who could break their security.
Later IBM revised CP67 and called it VM, and since then it has been from common to standard for the IBM mainframe operating systems to run on VM instead of on the bare metal. There have been decades of high end production systems running on VM. I haven't heard about any security holes.
As of a few years ago, a list from Microsoft of security holes fixed included at least one based on a "buffer overflow" bug. Gads: If as recently as a few years ago Microsoft still had buffer overflow bugs, one has to question if by then they much cared about computer security at all. Buffer overflow bugs -- that's Programming 101 for middle school.
I just checked Firefox 35.0.1 and couldn't find where to turn off JavaScript. For the Web pages at my Web site, I have so far not written a single line or character of JavaScript and hope never to, although Microsoft's ASP.NET does write some for me -- and I do wish I knew what ASP.NET classes or options I used to cause Microsoft to write any JavaScript for me.
The old first rule of computer security was to separate code and data, and never but never let data from an untrusted source run as data. Never. Not once. For any reason.
So, in light of this first rule, we have JavaScript downloaded with the HTML text and markup of a Web page and executed. So, I want a very, very, very clear, careful, rock solid, expertly reviewed, as close as possible to proofs of correctness evidence that it is impossible, absolutely, positively totally, without any possibly of exception, for code, any code at all that could possibly exist, in JavaScript that could cause my computer any problems at all, and otherwise just block it. Nothing to do with it. Dump it.
Computer viruses have cost me about half my time so far this year, and I want nothing to do with more computer viruses. JavaScript? Totally glad to junk it.
As I recall, not so long ago, Firefox had Java enabled by default! Outrageous. And at times in Microsoft's Internet Explorer had to be careful to disable Active X, which could run any code at all.
JavaScript? I want nothing to do with pull downs, pop-ups, roll-overs, the screen jumping around for no good reason, etc. Just HTML and CSS -- fine with me.
HTML was just a word-whacking mark-up language -- fine. But, yes, for user input it has text boxes, multi-line text boxes, check boxes, and radio buttons. Okay. Should be able to implement those safely enough.
Can we start to take computer security seriously? When are we going to start?
E.g., Microsoft has gone from Windows NT to 2000, XP, Vista, 8, 8.1 etc., but where in there is the solid security? They have the Start button, remove it, put it back, the Metro interface, tablets, phones, etc., but what about computer security?
I don't give even one weak little hollow hoot about touch screens, but I care a lot about computer security.
CP67 ... security provisions appeared to be
absolute -- run any code at all ... with full
safety and security.
There have been decades of high end production
systems running on VM. I haven't heard about
any security holes.
Once upon a time, perhaps 40 years ago (it's been very long, my memory is hardly exact) some IBM employees/researchers started with either CP67 or VM source code (I forget which) and found literally dozens of bugs and/or usable exploits for it.
Source code was not an unreasonable starting point, since IBM published all this code on microfiche, and also probably on mag tape.
IIRC many/most of the exploits revolved around emulating the I/O channel architecture, channel programs, and corresponding SIO or SIOF instructions.
They wrote the whole thing up in a prestigious publication, perhaps the IBM Systems Journal or Communications of the ACM or maybe ACM Computing Surveys. I did a very quick search but couldn't find that particular discussion.
Edit: I may have found the original paper. First I found:
A Taxonomy of Computer Program Security Flaws, with Examples
https://cwe.mitre.org/documents/sources/ATaxonomyofComputerP...
in which they reference a 1976 IBM paper that discussed many VM vulnerabilities. Here's just one of multiple times they cite that paper:
Case: I2
Source: C.R. Attanasio, P.W. Markstein, and R.J. Phillips, ‘‘Penetrating an operating system: a study of VM/370 integrity,’’ IBM Systems Journal, 1976, pp. 102-116. System: IBM VM/370
Description: By carefully exploiting an oversight in condition-code checking (a retrofit in the basic VM/370 design) and the fact that CPU and I/O channel programs could execute simultaneously, a penetrator could gain control of the system. Further details of this flaw are not provided in the cited source, but it appears that a logic error (‘‘oversight in condition- code checking’’) was at least partly to blame.
Of course, the first thing someone will write for it is LISP, thus invalidating your "old rules of code" by using "old code design".
These days, one can run OS from an FPGA without impossible effort. If I had significant financials at stake, that's what I would be doing.
I have had no losses, time or money, from viruses, worms, trojans or otherwise in my 25 years of computer use. I must be just lucky.
And because the attack surface was much smaller: No TCP/IP, DVDs, thumb drives, etc.