Slashdot Mirror
Netscape Founder Says Web Browsing Innovation Dead
mattOzan writes "Marc Andreessen told Reuters today that browser innovation ended five years ago (which would put us at about Navigator 4.5 beta -- what was so innovative about that? The "What's Related" button? Beatnik integration?) "Navigation is an embarrassment. Using bookmarks and back and forth buttons -- we had about eighteen different things we had in mind for the browser." Well, pass me the NDA and tell me what they were!"
It just happens to coincide with the time he left Netscape to go start his own failed company LoudCloud.
5 years ago was a great time, though. Good times.
I have been pwned because my
What about tabbed browsing and mouse gestures? Opera is still innovating with dozens of features. Now if only pages would render properly on it.
The truth is, that Netscape stopped all innovation at 4.5.
The rest of the world moved on, and they STILL don't see that.
Bookmarks, back and forward buttons are FINE, the real innovation is in the content, and the display of said content.
CSS, Macromedia Flash, PHP, etc are all great web innovations that continue to push the envelope.
Just because natural selection weeded out netscape doesn't mean the rest of the world stopped innovating.
I would expect such blatant racism on Fark, but on Slashdot? Mods please ban this asshole.
Microsoft have got the market, they don't need to do any work to keep it, so why add furthur inovations to IE, no reason at all, theve even held back on full PNG support, well the work doesn't need to be done so why do it?
And everyone emulates IE....
Well, I'd say that the browsers actually adhering to standards instead of doing whatever they feel like seems like an innovation... of course, adhering to standards means you can't implement every bright idea you get, so yes, it slows down the rate of change.
"Freedom means freedom for everybody" -- Dick Cheney
"Navigation is an embarrassment.
I think what he meant was "Navigator is an embarassment."
Using bookmarks and back and forth buttons -- we had about eighteen different things we had in mind for the browser."
Well IE is sort of better at this, in that favorites are individual files, so you can use the filesystem's find function to search (nice when you have 1000+ bookmarks).
And I guess he hasn't seen Opera's gestures?
well, word processing hasn't changed all that much either in the last five years.
Really Mozilla is still available. If he has better ideas then he is still free to develop them himself or push others to do it. Browsing is a mature concept now. It doesn't need to constantly change.. that'd make it hard on users. If he has ideas though I'm sure people would listen.
At what price learning? At what cost wisdom? The price is a man's peace of mind, and the cost is his life.
Nothing ... I think that's the point ... that was the first evidence of stagnation. Compare NS4.5 navigation to Mozilla 1.4 navigation to MSIE 3 navigation to MSIE 6 navigation and you're stuck with essentially the same model for all this time.
... damn I am trying to remember the earliest and I can't ... it started with a "D" or "Q" and was developed by the folks who made a very popular BBS terminal program ... humbug, sorry, I usually like to have my facts in line but the memory is fading) so while it is VERY nice, it's not truly new. And pop-up suppression isn't an aid to navigation, but a method to sanitize the code from the remote site.
And before people jump up and down about CSS and XHTML, remember that Andreesen was talking about browser navigation not layout technologies or other areas that are dominated by W3C.org.
I will mention that I think tab based browsing and the suppression of pop-ups have been two major boons to my browsing. However, I saw browsers with tabs back before IE 2.0 had come out (back when non-Netscape/IE/Opera/KHTML browsers were often integrated with your Winsock communications stack
It is more productive to voice thoughtful opinions (reply) than to judge (moderate) others.
As far as I can tell, the web really hasn't evolved that much either(not counting the browser). We are using the same protocols and delivering basically the same types of information that we did 5 years ago. Sure we have flash and other funky plugins to spice things up, but the backbone is still the same. How are you suppose to innovate when the set of building blocks you have to work with haven't changed? Sure you can mix them up a little and get mouse gestures and tabbing and such, but you need new building materials to work with to innovate on top of. Once we migrate off the current version of protocols that we so fondly call the Internet, and open things up some more, I'm sure we'll start to see this innovation.
-my other sig is your mom
IIRC, there were no tabs in NS4.5, and tabs are the thing I enjoy most in modern browsers. Then there is the search fields in the toolbar, very cool, plus Opera's location bar prefixs, I love being able to type 'g innovation' or 'a domain:au news' in the location bar and have a google or alltheweb search come back.
And some of the innovation is coming from web page developers rather than the browser, some java applets are getting very nice. Robust, functional etc.
And then of course there is XUL, which is IMHO brilliant, but likely to die. To be able to turn the browser into another application with a markup language is way beyond cool.
In short, I think Marc is spitting sour grapes.
>>
I am the director, and this is my movie
IMHO he's right, although I don't think NS 4.5, was the cut off point for such innovation. What he's talking about is large and dramatic innovation, not add-ons and great expansions (like Tab's, Gestures etc).
But this isn't necessary a bad thing, everyone who uses the net is currently used to using a web browser and its heuristically defined layout, back, forward, reload, home and stop. It doesn't really need (currently) to be changed, the same applies to the controls of a car, the way a book works or even mobile phone interfaces. It works this way, billions of people use it such and changing it would have to be for dramatic purposes.
It doesn't stop us refining it though (again, Tabs, Gestures), just like a car (ABS, Sat Nav, Power Steering etc).
It may not be the best solution, but what about something like this: a 'teach gestures' option; when checked, every time the user did something another way that could be more efficiently done with a gesture, this would display a popup with a diagram of the relevant technique.
Except for, you know, some minor things like browser support for the DOM (which is huge), CSS (which is huger), XML, XSLT, and XHTML.
Or maybe he's just talking about the UI side, where we've seen absolutely no improvement whatsoever. Except for tabbed browsing, mouse gestures, integrated search bars, and popup blocking (though back then, popups weren't so much a problem).
Which is to say, really, that he's wrong. Sure, browser development is arguably slower now than it was back in the Navigator 1.2 Perpetual Beta days, but that's always the case -- the mad rush of innovation has to slow down after the low-hanging fruit is plucked. It certainly hasn't stopped, though.
that using a webpage requires memorization. The widgets on a webpage are quite different than those in most GUIs. They are HIGHLY customizable, and not necessarily themed to match the rest of the OS. This causes major conceptual problems for those unused to computers. The webbrowser is a totally different UI in many ways than the rest of the computer, a UI that can only be learned by memory.
Photos.
Yeah, and innovation for the book died when they created the index, the table contents, and page numbering. As long as the glue that binds the book holds and the ink doesn't run when it gets wet, I'm happy.
<a href="http://www.joblessjimmy.com">Work is dumb and so is Jobless Jimmy.</a>
Well IE is sort of better at this, in that favorites are individual files, so you can use the filesystem's find function to search (nice when you have 1000+ bookmarks).
Oh, I hate that one-file-per-bookmark idea. You aren't allowed to call the bookmark whatever you want -- why did they disallow characters like '?' or ':', instead of BASE64-encoding them or something? And these days it's not so bad, because most people are running FAT32 or something better, but back in the day there were a lot of people running FAT16, and on a 2GB disk partition, each bookmark used up 32KB of storage! Yikes!
I'd rather just have a non-sucky UI for finding inside the bookmarks file. (I've just started using Mozilla Firebird and so far the bookmarks searching seems pretty good.)
steveha
lf(1): it's like ls(1) but sorts filenames by extension, tersely
well I'm glad he thinks browser innovation is dead. now how about they start working on properly supporting things like CSS!
So incredibly annoying building a page to perfect standards and having a browser munge it anyways!
A year spent in artificial intelligence is enough to make one believe in God.
Radial Context Menus + Tabbed Browsing Extensions for Firebird (Mozilla) browser are the most radical leap forward in surfing efficiency since the invention of the browser.
Nothing has come close to increasing speed, efficiency and general surfing pleasure as these two items, combined and tuned to perfection.
Andreesen is obviously majorly deluded, and the publishers of the article are obviously morons.
If there were no Flash, there would be no Homestar Runner. And that would truly be a sad thing.
Browser integration is not dead! Microsoft works on it everyday. MS constantly innovates new features that crash Netscape, Mozilla, and Opera but provide no value. Innovation will truly end when we can no longer use any browser other than IE and at the rate things are doing it won't be long.
... ) As you can tell I am not the least bit bitter. I hate tools that generate bloated, crappy, IE specific HTML laden with self destructing Java script.
I would ask all the developers out there to support more than IE on your extranets. I am talking to YOU Mr. Webdeveloper in that Fortune 100 company like Ford, GM, Diamler, EBAY (Is Ebay Fortune 500 yet?). Ask yourself this, "Do I really want to limit the web to a Microsoft ONLY browser?" The point of the web was platform independence. I especially love developers who code in Java then create an O/S specific dependency.
Think about this next time you decide to implement a feature that only works in IE but provides little to no value to the end user experience. If we all wake up one morning and find we are living in a "one-browser", "one-platform" world, it is going to be horrible. However, if it makes you feel any better, I fight these battles every year and lose to the developers. Usually the management will wine and say, "it costs too much to develop for two browers" or "but if I can just let developers wiggle the mouse and use a tool to generate HTML they won't have to think and can get my project done faster".
However, web browers in general suck for application development. I think the old mainframe character terminal had better input screen capability than the modern web browers. In fact, if you compare the two they work just about the same (Push screen to terminal/browser from, Fill In Screen, send screen back to server, repeat
The whole "embrace and extend" concept is getting old. Can't we all just get along and make things better instead of creating a fragmented incompatible mess. A company like MS is sitting on piles of cash, the likes of which the world has never seen and instead of putting together a Bell Labs or Xerox Parc they spend their cash on marketing and lawyers. There are so few great men in our age and even fewer visionaries. I was holding out a glimmer of hope for Bill when he took over as chief software architect. So much for Utopia.
"There is nothing new under the sun." -Solomon
Grandparent is missing the point -- it's not that browsers should not have forward back and bookmark, somehow replacing these with "better" functional metaphors. It's that they shouldn't have only forward, back and bookmark as navigation and information storage paradigms.
Parent, on the other hand, is on to it -- keep going! How can browsers make it easier to remember what we saw, where we saw it, and why we cared when we saw it? These are the questions that don't seem to be influencing modern browsers (or any browsers, ever, see below).
Gestures are cool, but they are functional improvements (immediate, operational efficiency enhancements) -- where are the higher-level, conceptual, long-term efficiency enhancers? Why can't my browser warn me if there is a reputable opionion on an arguable topic I've been researching that I have not yet read? Google knows, or very nearly does. It's a challenging but possible leap to make a browser be able to understand sets of info (refer to Google sets, google for it if you don't know, it will blow you away. It basically takes a few items from you, figures out what is in common between them, and fills in the other things like them. All from web context clues.) Why doesn't my browser note that I'm checking out info on items A and B, look up the fact that these are both items in set z, and then gently suggest that I also check out the other items C, D, and E since it knows these are also in set z? Maybe they're all in set y too, offering yet another angle -- the browser should know. My point is that the info should be there, making it available inobtrusively is a trivial detail for interface designers to ponder.
I can do more with perl and wget (or LWP) in less time than any browser that exists, and I do occasionally resort to that when searching a tricky topic. This should not be a true statement.
Why can't my browser (at least pretend to) understand some of the info I see every day, categorize it, and make sure I can find it (and extract summaries from it) later, easily? The technologies exist (data mining, xml, bayesian filters, crude ai) but they have not been integrated into browsers. Tivo lets you thumb up/down any content and thus vote your preference to see more of the same. Why don't browsers have something like this? (To be fair, I have seen attempts at this, but they all tend to degenerate into advertising-ruined information dead-ends.) And why can't it learn (or ask) why I did/didn't like a site, and extract from that aggregate info what sites I might like or not (maybe even including some % of what my friends like.) Then from this form bayesian-like filters (more intereactive than those used for emails these days) to help prioritize (not filter, really) data. I'm thinking of a meta-google appliance that applies your own categories of interest and weighting preferences to google pagerank results, re-ordering the results for your preferences (i.e., I am a member of the European Demolition Association, so searches for 'EDA' should show me demolition-related hits before Electronic Design Automation hits, which would otherwise dominate the first-listed google results.)
Let me steer you a bit more another way -- it's what we have seen that's important. Google is doing a great job of letting me find new stuff. No problems there, but what happens when you need to go back later and find that really cool site on that topic that just happened to come up again a few days later and ooh, it was so relevant and full of meaty info and if I could just rememeber the keywords I used to find it . . .
So, css, gestures, etc. aside (they are innovations, but minor, and not involving any major architectural change), we haven't see much change or innovation since the very first browsers created. Other than speed, some standards changes, and aesthetics, you can use Mosaic 1.0 to find info on the www pretty much as easily as IE or any of it's modern competitors.
And that's the point he should have made.
everything in moderation
Take a look at The Brain for an innovation in browsing. I'd like to see more sites adopt this sort of navigation scheme. Something that's always bothered me about browsers (I use IE primarily, as I'm one of those unfortunates that is locked into Windows) is the disgusting underuse of the "Forward" button. I don't know how many times I've backed up on a path, gone down some other path, then wanted to get back to where I was. I could back up to the fork point, but didn't have any "Forward" options other than where I just came from.
The Spoon
Updated 6/28/2011
To argue that innovation died back in 1998 is to ignore much of the progress Opera and Mozilla have made and continue to make. Both Opera and Mozilla offer new features like tabbed browsing and pop-up blocking. No, there hasn't been anything earthshattering, but we have seen many refinements of the general idea. The refinements are significant enough, in my opinion at least, to make Internet Explorer or even an old Netscape Communicator seem primitive in comparison. Andreessen's funny assertion that innovation is dead reminds me of a story I once heard (and whose truth I question, but I digress) about some European country that, towards the end of the 19th century, had a government that voted on closing down its patent offices. Why? Because everything that would ever be invented had already been invented. Of course, we would look back on that and think it is absurd. To me, Andreessen's logic is really no different - it is bullish and stubborn. Or, as another reader so aptly put it, bitter.
Yeah, and I *hate* that. 90+% of the PDF documents I come across could have been done just as well in HTML, where the user has control over font size and the text isn't artifically constrained to a "page" view which makes no sense whatsoever when reading on a monitor.
How to solve most of our problems: 1.Lots of nuclear plants. 2.Cure aging.
Wow, this isn't well thought out at all.
Javascript, more properly DHTML, is amazingly better than it was at the height of the browser wars. Compare Danny Goodman's DHTML Definitive Guide 1st and 2nd editions. The first one is all about how to handle the differences between Netscape and MS. The 2nd is all about documenting the *standard* DOM and how to script it. MUCH easier than it was 5 years (or 3 years) ago.
Java applets are slow -- no argument there, but hello? Sun? Java is controlled nearly as tightly as Flash. And if you read Macromedia's marketing, they're basically presenting Flash as the Visual Basic of the web. Love or hate VB, it certainly made it easier to build applications. Something similar for the web would have a similar effect -- increased ability for developers to write great apps and increasingly crappy code from non-developers who think they're developers since easy GUI tools lower the barrier to entry for development projects.
Abuse of the technology -- cookie abuse, popups, etc -- is not the fault of the browser. tech is tech, use is use.
And I'm no fan of Microsoft, but Internet Explorer is a *great* browser. Again, wind the clock back 5, or even 3 years. And 6.x is now the last of it's kind before MS rolls it back into the OS.
Exactly. I would much rather see browsers considered mature technology while getting their standards correct then more tacked on 'enhancements.'
I don't want a browser that's secretly a P2P app.
I don't want embedded media and plug-ins crashing it.
I don't want a browser that is also a PIM.
I don't want a little avatar asking me if I want to go to shamelessmarketers.com.
etc.
Why does everything have to be attached to the browser? A simple interface and a stable platform is what companies should be aiming for, with the exception of tweaks and minor enhancements like pop-up blocking, tabs, etc.
The Mozilla team has learned from this mistake. People kept complaining about the "Mozilla Suite" and the bloat and they responded by announcing plans to seperate the browser from the suite.
Microsoft in the meantime continues its "the browser is the desktop" nonsense which mixes WAN data with the OS. As we've seen with ActiveX, vbs, etc this is a security nightmare.
I'm not sure what Andreesen was secretly planning, but an url box, back/forward buttons, and a stop button are surprisingly effecient when dealing with html-based technologies.
I'm giving up my mod points so I hope the other moderators will mod up the parent.
I cannot agree enough with your post. I just finished an application written entirely in DOM, CSS, and Javascript. The HTML frames are generated entirely out of Javascript code. No "regular" HTML is sent to the client.
I kept to bare W3C DOM objects and methods, such as addChild, document.createElement, and so forth. Guess what?
It works in IE 6+, Mozilla (+derivatives), Safari, and others. No browser detection. No special coding. No hacks.
Also, note that this is a full blown web-based application so I feel justified in asking my users to upgrade their browsers. I wouldn't do this on a home page or regular site that people visit. Eventually we can expect 6+ browsers for home pages, but not yet.
Also also, despite my thinking the app is pretty cool in its dynamic interfaces, I can't say enough how much of a screwed up language Javascript is. The companies have really screwed us this time. It's a pain to debug. It's a pain to write (being combined with another server-side language, python in my case).
It's too bad that I think DHTML is the future. I really do think it will make it because it achieves dynamic content without plugins. I just wish it was cleaner. Perhaps IE will finally suppor W3C standards and the language/DOM support will clean up as time goes on. I'm hoping but not holding my breath...
I can't believe that was modded as insightful. UNIX tends to use a single configuration file for a single purpose. You rarely see a 100+ files to manage an application. The bookmark file for a browser is just a configuration file. When MS moved to make the web browser the file manager, they decided to cut out the code for managing bookmarks, and let the file management portion of the code do the work. The windows registry is not a unique concept. Just check out AIX, and you will see the ODM database is very similiar(don't I remember an old partnership between IBM and MS. They couldn't have stolen the idea, could they have?). The difference is a corrupt ODM will not prevent me from booting the machine. It can be rebuilt in single user mode at the very least, and the applications still store thier own settings. A horked windows registry equates to a useless windows system.
Extremely long work hours.
No clue what your company was really about and how it had any hope in hell of ever turning a profit someday.
Loads of sleazy people in the industry.
HUGE egos everywhere (dot-snobs).
Impossible to keep up with all the latest and greatest "next hot things".
Everybody spouting off like they know everything.
John Kerry is a Joke!
If the things they "had in mind" were anything like their context-destroying Frame model, or their DTD-breaking Object substitute Embed: Good riddance.
Why didn't they implement proper support for Link relationships? Why did they feel the need to make their own Java security model? Why did they hack their own Javascript-based styling instead of just implementing CSS properly?
The software industry is better off without them. A worse case of "Not Invented Here" mentality is hard to find.
The thing Marc Andreessen does not say is that all the innovation is not around HTML anymore. It's RSS, Echo (well, soon :), two way communications in Blogs (Trackback, Pingback, Referrer lists, etc), FOAF, GeoURL, etc.
For the moment, all these higher level ideas are being integrated into web pages, because the browsers aren't using them (except for RSS readers).
Today's browsers are the user interface to HTML. We still have to invent the user interface for these technologies. They are the next layer of the web.
I mean, this guy and his team basically took a horribly broken tagsoup interpreter and added proprietary extensions to it. It was certainly an important step in the evolution of the WWW from a low-tech hypertext information system to a distributed advertising platform, but I fail to see why he should be met with any kind of respect.
Programming can be fun again. Film at 11.
You want dynamic pages? Have the browser call a C++ or Java program binary directly and screw all this other crap...JavaScript, Java applets, Flash, Perl scripts, Jesus, what a nightmare...
You mean like ActiveX? That way I can make a program that does anything I want (including destroying all your documents and software and doing my best to take down your hardware), because as long as I'm executing intel assembler instructions I can always break out of any attempted sandbox. ActiveX programs are precompiled programs that your browser happily downloads and executes for you. I LIKE the fact that java applets are sandboxed. I LIKE the fact that javascript is limited in what it can do. But you want web page developers to be able to excute any code they want on your computer?!?
I guess he was lucky to be in netscape at the time, most of what he did afterwards kind of failed miserably.
/. is probably an idea of questionable worth.
I know people from NCSA who knew Andreesen fairly well. The guy is no great oracle/wunderkind. He just got lucky to be in the right place at the right time. The rest was all marketing by Netscape to try to push the value of their company.
I'm not trying to put him down or anything -- I'm just saying that posting everything he says to the front page of
Why are you letting these clowns ruin our country?