Choose Theme

Find the Sexy Nerd Next Door? 

News & For Sale “Firefox Is Ten Years Old: Happy Birthday!” by m1cnBot

 
Member
m1cnBot 5 years ago on 09/24/12

Story Originally Posted By: John Booty

Article PictureThere wouldn't have been an OtakuBooty without it. Firefox saved the World Wide Web.

The very first version (called Phoenix) was released to the public ten years ago today.

Incidentally, give Firefox a try if you haven't checked it out in a while. Still my favorite browser, and the last few releases have been lean and snappy.

item
Member
John Booty 5 years ago on 09/24/12
Equipped: Sparkledonkey's Gallbladder
One good reason to use Firefox in 2012 is that Mozilla is a non-profit corporation. They exist pretty much just to make the web better.

Chrome beats Firefox on some technical levels but not others. However, the thing is, Google not exactly your friend. Google creates software like Chrome because today (and for the foreseeable future) Google most certainly believes that a free and open Internet is crucial to their business plans.

So that's nice, and we get a nice free web browser (Chrome) as a side effect of that.

But that's not Google's primary mission. They are a public, for-profit company, and their primary goal is to make money for themselves and for shareholders. I worry about how their desire for a free and open Internet might change if/when they achieve browser market share approaching what IE once had.

Complicating the issue, of course, is the fact that Mozilla receives a large percentage of their funding from a contract with Google for making Google Firefox's default search engine. However, they're not necessarily totally dependent on Google - if Google or Mozilla ever discontinue that contract, Mozilla can of course turn to other providers.
item
Member
John Booty 5 years ago on 09/24/12
Equipped: Sparkledonkey's Gallbladder
Looking to the future, did you know that Mozilla is working on a Firefox mobile OS? Devices are rolling out soon.

http://rawkes.com/...is-something-magical-about-firefox-os

I think this has the chance to be pretty important. I don't think it's going to ever be some kind of dominant mobile OS or anything - it's not conquering iPhones or even mainstream Android phones.

But it seems like it's going to be the most hackable mobile OS - the entire thing is based around web standards. If you can make a web page, you can make an "app" for something running Firefox OS.

Obviously, Palm/HP tried that with Web OS or whatever. People liked it, but it died in the marketplace without support. Firefox OS's economics are totally different; it's not something that needs to make hundreds of millions of dollars to live. It can build momentum slowly.

I'd loooove to have a tablet running this.
item
Member
Kumba 5 years ago on 09/24/12
Equipped: Most Amazing Thing Ever!!!! named "Silmarils, gems of treelight"
Mozilla needs to knock it off with the version race w/ Chrome, and put their UI developers in the backseat and focus on improving other areas of FF rather than constantly trying to force UI changes down user's throats.

I.e., there is this bug that I myself opened back in FF9 that they STILL haven't fixed yet, despite basic acknowledgement by the guy who introduced the regression that it's a problem:
https://bugzilla.mozilla.org/show_bug.cgi?id=712763

Still, despite the flaws FF has, yeah, it's the best browser out there right now because of its extreme flexibility w/ the add-on system. Still, sometimes, some of the developers decisions leaves you wondering and/or scratching your head. I really should check out Palemoon at some point, or just try to switch to Seamonkey.
item
Member
John Booty 5 years ago on 09/24/12
Equipped: Sparkledonkey's Gallbladder
That's like telling your pizza delivery guy to drive slower on his way to your house so that they can make better pizza. The quality of the pizza is not related to how fast he drives.

The rapid release cycle allows features and fixes to reach users sooner. It's not something they do instead of developing those features!

Now that FF updates silently, like Chrome, this will hopefully become the non-issue in peoples' minds that it ought to be.

I really should check out Palemoon at some point


I ran Palemoon for a while. I recommend checking it out; it uses your Firefox profile so there's very little overhead in switching to Palemoon or switching from Palemoon back to Firefox.

I switched away from Palemoon because 1) a lot of the speed gains are just things you can do with about:config anyway and 2) last I checked, the Palemoon guy stripped developer tools out as part of his "slimming" process. I would obviously rather have those.
 
Member
A Study in Pink 5 years ago on 09/25/12
I was a die hard ff fan, but I had to switch because my browser was crashing every single day, usually due to memory leaks from a poorly designed flash application. That was about a year ago, I think. One of those rapid release whatevers just completely made ff unworkable for me on all of my machines, ubuntu and osx alike. It just doesn't do that in chrome. Whether or not it's fixed now is pretty irrelevant because I've got my extensions just the way I want them in chrome and I have no practical reason to switch.

I'm not passionate about the ideology enough, but my love for Google definitely died when they reopened their site in China. Oh, and this bullshit about predictive search results. A lot of things, actually, have made me much more suspicious about our "benevolent" internet overlords, but I see no personal gain for switching back to ff.
item
Member
John Booty 5 years ago on 09/25/12
Equipped: Sparkledonkey's Gallbladder
The main feature of FF15 (the current major version) is that extensions are now prevented from leaking memory. I'm not sure if there's anything like that in place for plugins, though.

There are definitely things I like better about Chrome. The biggest one is how nice and integrated the developer tools are. Firebug is nice, but kind of a sloppy mess next to Chrome's tools. I think FF is gradually heading in the direction of a more integrated set of dev tools, but they're a long way away from supplanting Firebug.



item
Member
WWNSX 5 years ago on 09/25/12
Equipped: Inappropriately Large Handful of Viagra
John Booty said:
I switched away from Palemoon because 1) a lot of the speed gains are just things you can do with about:config anyway


Before I google do you have a link to a list of those modifications? If that's the only difference then I rather just go into about:config myself.

Thanks Kumba for making me think of Netscape Communicator. I lost so much past e-mail due not being able to export stuff from NC into a different applications especially since it was a all in one application.

I find Thunderbird to be the best for e-mail more then any other e-mail client, so I hope it keeps on going. I rather not be stuck in a all in one application ever again.

If you're running FF I still think Memory Fox is a mandatory extension as it helps with firefox's memory leaks, memory hogging tendency, and things like flash. Yes even with the current version.
item
Member
John Booty 5 years ago on 09/26/12
Equipped: Sparkledonkey's Gallbladder
Before I answer that question, let me say that I follow various Mozilla employee blogs, mailing lists, etc.

Obviously that doesn't make me an expert but I do have a sense of... which way the wind is blowing. What I can say is this: Mozilla takes performance really seriously. They care a lot, and spend a lot of time working on it.

So, please don't think there are huge performance gains sitting around in about:config waiting to be turned on. If there were, Mozilla would have flipped those switches already. In general, you are very likely to make things worse with alleged about:config performance tweaks.

Okay?







That said... I know "network.http.proxy.pipelining" which is false by default in stock Firefox and true by default in Pale Moon. I believe this is also true for "network.http.pipelining" but I'm not sure.

This page indirectly mentions some of the about:config tweaks.
http://www.palemoon.org/troubleshooting.shtml#http_proxy

The reason they're false by default is because pipelined http requests fail on some reallllly old/shitty web servers and such and browser makers such as Mozilla (the others do it too) don't want to essentially make those sites unavailable.
item
Member
John Booty 5 years ago on 09/26/12
Equipped: Sparkledonkey's Gallbladder
Also, OS memory management is a complicated issue. The short version of the story is that, if you think Firefox is using "too much" memory because you opened up Task Manager and saw that it's using eleventy gajillion jiggabytes (and think that you "helped" things by using MemoryFox) just please know that the Task Manager's numbers are an extremely crude at best and completely misleading at worst snapshot of what is going on with Windows' memory at that moment.

And there is no place you can go for some kind of concise numbers because the actual answer is "it's complicated" and it involves like an entire 200- or 300-level course learning about things like working sets, public/private bytes, the system cache, etc.

The best metric is "how is my computer running? does shit feel snappy?" and not the Task Manager numbers. Forget those.

From a guy who has spent his life learning about this, do not look at anything in Task Manager besides maybe the CPU graph.
PS: That is true for any modern OS: Windows, OSX, Linux, whatever.
item
Member
nategri 5 years ago on 09/26/12
Equipped: Clue Stick named "Some Sense"
Been an on and off user since '03. Can't believe it's been that long.

I think my browser use history goes something like [IE (It was the late 90's -- shove off!)] -> [Netscape] -> [Mozilla] -> [Firefox] -> [Opera] -> [Firefox] -> [Chrome]

Currently I have all of the last three installed, but Chrome is my go-to.
item
Member
WWNSX 5 years ago on 09/27/12
Equipped: Inappropriately Large Handful of Viagra
John Booty said:
Also, OS memory management is a complicated issue. The short version of the story is that, if you think Firefox is using "too much" memory because you opened up Task Manager and saw that it's using eleventy gajillion jiggabytes (and think that you ...


Well yeah that's true if you're not selecting the other view columns but I still use it as a general guide and not this is what the actual program is using memory wise since that will always change.

Flash is still just as bad as firefox though which is still a issue of course. for me memory fox still works for my 50 to 200 tab habit, so I use it on any and every system.

Thanks for the link and info on what it does. I'll see if I can find more.

I'm not necessarily concerned about the speed tweak but it's nice knowing what they are and what they do.

my early browser history went like this: Mosaic -> Netscape -> Firefox

If I even used IE it was to get Netscape or Firefox.
item
Member
John Booty 5 years ago on 09/27/12
Equipped: Sparkledonkey's Gallbladder
Mine was:

Viola (in 1994, I think) --> Mosaic --> Netscape 3 --> IE 4 --> Opera --> Mozilla --> Firefox

item
Member
Kumba 5 years ago on 09/30/12
Equipped: Most Amazing Thing Ever!!!! named "Silmarils, gems of treelight"
John Booty said:
The rapid release cycle allows features and fixes to reach users sooner. It's not something they do instead of developing those features!

Yes and no. This development cycle, if used properly, would be better. But right now, FF development has fallen into the same trap as Gnome 3 and Windows 8 -- UI developers have hijacked things and are developing features that they think the users want, rather than listening to feedback users.

Reference the FF bug I linked above -- The developer who added it only tested it on Mac OS X, which allows you to re-arrange program groups fairly easily, form what I gather. But on Windows Vista and earlier, if a window takes too long, Windows will move that taskbar button to the back of the taskbar and get stuff out of order. On Windows 7 and up, you can only move the entire group, but individual buttons within the group. So when shutting down FF, you have to have focus on the window that you want to be at the front of the group.

That's a regression that could have been easily avoided had that Mozilla developer tested it on other OSes. It's just one example of the sloppiness that's crept into Mozilla's development cycle as-of late.

Also, I cite this statement on their webpage for the ESR releases (https://www.mozilla.org/en-US/firefox/organizations/) (emphasis mine):
Who is it for?

Firefox ESR is intended for groups who deploy and maintain the desktop environment in large organizations such as universities and other schools, county or city governments and businesses.
Who is it not for?

Individual users who always want the latest features, performance enhancements and technologies in their browser without waiting for them to become available in ESR several development cycles later.


Those three words are indicative of the arrogance that's taken over the Mozilla development cycle. They assume all users always want the latest and greatest, and that is not always the case. As such, they make you hop through several links just to get to the ESR release because of this assumption.


WWNSX said:
Thanks Kumba for making me think of Netscape Communicator. I lost so much past e-mail due not being able to export stuff from NC into a different applications especially since it was a all in one application.

This is still pretty hard to do in Thunderbird, unless there's some add-on I don't know about. It's easier if you're just moving data from one Thunderbird profile to another, simply copy the contents of your profile into the new one on the target. But it is a far cry from Outlook's Personal Folder Files feature that lets you archive off stuff into PST files that can then be backed up somewhere else.

The upside is OpenPGP/GnuPG integration is better than with Outlook if you have Enigmail.
item
Member
Kumba 5 years ago on 09/30/12
Equipped: Most Amazing Thing Ever!!!! named "Silmarils, gems of treelight"
John Booty said:
The reason they're false by default is because pipelined http requests fail on some reallllly old/shitty web servers and such and browser makers such as Mozilla (the others do it too) don't want to essentially make those sites unavailable.

This is really the fault of TCP forcing HTTP to include data streaming features, because TCP's "here's a big bucket of assorted bytes" approach makes independent data streams all but impossible. Multiple data streams REALLY should be handled at the transport level so that the HTTP layer isn't cluttered up with half-assed implementations of pipelining across various client/server products that sometimes aren't compatible with each other.

It's a pity people would rather continue developing and enhancing this kind of brokenness instead of switching to a more robust transport protocol like Stream Control Transmission Protocol (SCTP). Independent multiple data streams, multi-homing (i.e., tying multiple IPv4/IPv6 addresses together within an "association" to provide automatic failover if one IP connection goes down), actual delineation/framing of the SCTP payload, and a bunch of other features make it pretty awesome. But I digress.


John Booty said:
And there is no place you can go for some kind of concise numbers because the actual answer is "it's complicated" and it involves like an entire 200- or 300-level course learning about things like working sets, public/private bytes, the system cache, etc.

I believe "Memory (private working set)" is the best column for gauging current memory use in Windows. I've identified memory leaks in programs before by monitoring this statistic (i.e., Pidgin will leak ~20MB a day if you chart this figure over a week or two).


WWNSX said:
Flash is still just as bad as firefox though which is still a issue of course. for me memory fox still works for my 50 to 200 tab habit, so I use it on any and every system.

Adobe should be tried and executed for crimes against humanity for continuing the Flash debacle after the Macromedia acquisition. Want to know why the information security/information assurance market has plenty of job opportunities? You can thank Flash (and several other Adobe products), ESPECIALLY authplay.dll, for the near-continuous stream of security vulnerabilities that keeps people in that sector employed.

Seriously, if Network/Systems administrators just enacted an outright ban on Flash, and probably ShockWave and Adobe Air, you'd significantly reduce the number of security nightmares out there.
item
Member
John Booty 5 years ago on 10/01/12
Equipped: Sparkledonkey's Gallbladder


Kumba said:
It's a pity people would rather continue developing and enhancing this kind of brokenness instead of switching to a more robust transport protocol like Stream Control Transmission Protocol (SCTP). Independent multiple data streams, multi-homing (i.e., tying multiple IPv4/IPv6 addresses together within an "association" to provide automatic failover if one IP connection goes down), actual delineation/framing of the SCTP payload, and a bunch of other features make it pretty awesome. But I digress.


Chrome and Firefox support SPDY, which runs on top of TCP and addresses some of this. HTTP over SPDY gives you SSH everything, pipelining, etc.

Mozilla said:
Individual users who always want the latest features, performance enhancements and technologies in their browser without waiting for them to become available in ESR several development cycles later.


Kumba said:
Those three words are indicative of the arrogance that's taken over the Mozilla development cycle. They assume all users always want the latest and greatest, and that is not always the case. As such, they make you hop through several links just to get to the ESR release because of this assumption.


Wow, dude, whoa. I'm really not reading that in remotely the same way you are! I don't think it's arrogant. I think their assumption that most users want frequent, nag-free updates is correct - and this is a totally reasonable assumption validated by Chrome's success with that model.

And then they provide the ESR releases as a non-default, alternative for people who just want "security fixes only; no new features." This approach has been really successful for Ubuntu.

Choosing a sensible default that works for most users isn't arrogance - it's like 90% of the art of designing good software.
item
Member
Kumba 5 years ago on 10/01/12
Equipped: Most Amazing Thing Ever!!!! named "Silmarils, gems of treelight"
John Booty said:
Chrome and Firefox support SPDY, which runs on top of TCP and addresses some of this. HTTP over SPDY gives you SSH everything, pipelining, etc.

SPDY's a cute hack at best :) It's essentially tunneling HTTP, as it's only implemented by browsers and, for now, requires a proxy-like daemon running on the server to convert to/from SPDY for the HTTP server. I don't think Apache nor IIS support SPDY natively. It's just another example of why this should have been implemented at the transport layer level, not the application layer.

SCTP, on the other hand, was developed by the telecommunications industry. And although few have probably ever heard of it, it's the backbone for most cellular networks, internal corporate VOIP, and even older SS7-based phone stacks. It needed to be robust, hence the features I listed earlier that would be incredibly useful for HTTP and other protocols. It's also one of your four IANA "general purpose protocols", along side TCP, UDP, and the quirky DCCP.

But that said, my argument is a moot point for SCTP because it's only implemented in Linux and FreeBSD, so while there are old patches for Firefox and Apache, they don't work anymore (and were really a TCP-over-SCTP kludge anyways).


John Booty said:
Wow, dude, whoa. I'm really not reading that in remotely the same way you are! I don't think it's arrogant. I think their assumption that most users want frequent, nag-free updates is correct - and this is a totally reasonable assumption validated by Chrome's success with that model.

Oh, their assumption probably is very correct. But I find there's a difference between being correct and how you phrasing that correctness. I find their phrasing to be arrogant, and the fact they hide the link to the ESR release down at the bottom of the FAQ page implies a degree of effort taken to hide the ESR release.

ESR was only created after public backlash to Asa Dotzler's comments that Mozilla doesn't care for the enterprise anyways. The rapid release schedule has made testing and validation by corporate IT and IA staff quite a burden. I pin the arrogance largely on him, and there's other statements of his out there that carry this as well.

http://...zdnet.com/...er-been-a-focus-of-ours-3040093220/
item
Member
Creepy Stalker 5 years ago on 10/01/12
Equipped: N00dz of akanepanda
:)
item
Member
John Booty 5 years ago on 10/01/12
Equipped: Sparkledonkey's Gallbladder
Kumba said:
I find their phrasing to be arrogant


OK. I think I see what's going on. I think you're inserting a comma in your mind or something. This would be kind of arrogant, because it presumes to speak for all individual users:

Nobody actually said:
Individual users, who always want the latest features, performance enhancements and technologies in their browser


But this is what they actually said. Note the lack of comma after users:

Mozilla said:
Individual users who always want the latest features, performance enhancements and technologies in their browser


They're not presuming all individual users want the latest stuff. They're saying that individual users who want the latest stuff can stick with the default (silent, frequent updates) and others can use the ESR releases.

Kumba said:
SCTP, on the other hand, was developed by the telecommunications industry. And although few have probably ever heard of it, it's the backbone for most cellular networks, internal corporate VOIP, and even older SS7-based phone stacks. It needed to be robust, hence the features I listed earlier that would be incredibly useful for HTTP and other protocols. It's also one of your four IANA "general purpose protocols", along side TCP, UDP, and the quirky DCCP.


Wouldn't this need to be supported at the hardware level? Like, everywhere? Switching to SCTP would have the ever-so-sliiiiiightly inconvenient side effect of not working with any currently available consumer networking gear.

Also, you know, just to digress from your digression... maybe the faults of TCP are also its strengths. Suppose that, back in the 1970s, TCP had included some of these more complex features like streaming, etc. That certainly would have made the protocol more complex, hardware likely would have been more expensive, and the added complexity raises the possibility of the real-world issue of incompatible implementations. Sometimes (not always) dumber is better.
item
Member
Kumba 5 years ago on 10/02/12
Equipped: Most Amazing Thing Ever!!!! named "Silmarils, gems of treelight"
John Booty said:
OK. I think I see what's going on. I think you're inserting a comma in your mind or something.

Trust me, it's not grammatical :) I am only citing that one statement as an example. Probably a bit of a weak example, but an example nonetheless.

This recent strain of the developers thinking for the users is not completely confined to just Mozilla. Take the recent spat in several Linux distributions over the /usr merge, itself instituted by a select few Fedora/Redhat developers, udev/systemd integration, the Gnome 3 UI change that irked quote a few people, and the can o' worms that Windows 8's UI is. It is a troubling trend, especially for open-source projects, when the developers say to the users "trust us, we know what you want".

Sure, there's a block of people out there that actually like that (laziness is the mother of all invention, after all), but the entire concept of open-source software is rooted in the idea that developers listen to the users and take their input into consideration. I mean, Linux would not be where it is today if Linus hadn't listened to feedback from people on the mailing list where he first published the code in 1991.

Also, I should add as a disclaimer that I am a bit of control freak and minimalist when it comes to software. This is why I favour Windows Server for a desktop because it's got all the bells/whistles disabled by default, not the other way around. Ditto for liking distros like Debian and developing for Gentoo.

You need a paid membership to reply to this thread.

Paid memberships are four bucks a month or twenty bucks a year. Cheap!

However, creating an account is free. Members without paid accounts can use a lot of the site's features without ever paying anything.

Be awesome. Create an account!

Geek dating and social networking for awesome people.

Sign Up. Join OtakuBooty!

OtakuBooty is where smart, funny, sexy nerds meet. Creating an account is free. Full membership is $4/month or $15/year. Cheap!

Press People. Need material? Cover OB for your site, blog, podcast, magazine, or what-have-you.  More info »

Want Your Stuff Reviewed By OB? Just send us your press releases and requests to review your products.