I have just checked in Bug 378216, and wanted to give a quick heads up on it.
What this means is that we are now enforcing a security restriction on all add-ons. To be specific, if an add-on does not provide a secure method of auto-updating then by default Firefox will refuse to install the add-on. If you have add-ons already installed that are insecure in this way then they will be automatically disabled.
The good news is that addons.mozilla.org already uses SSL for it’s updates, so any add-ons you have installed from there will be unaffected by this change. Equally any add-on authors who use SSL on their site, their add-ons will be unaffected. Personally I found 2 of my add-ons were disabled by it, that’s 2 out of nearly 20, so hopefully you won’t see a major impact.
For add-on authors there is an alternate way to provide secure updates without investing in an SSL key involving digital signatures, unfortunately we’ve had to hold off on providing the software to make that possible until the backend changes were complete and reviewed. I hope to have something usable available not too long after M8 is released. Keep an eye on this blog for an update on that.
If you want to see more of the specifics the best place to look is probably at the wiki page. This is all based around the discussions I started on various forums and newsgroups. Hopefully it’s not too much of a surprise to the add-on authors out there, if it is then I apologise, I tried to get the word out as best I could.
After a fair bit of work (feels like longer than 2 months) I’ve finally managed to get bug 382752 landed. What this gives us in simple terms is a set of functions that we can use in order to do unit testing on the extension manager. Alongside I have checked in the first unit test. Now if anything regresses bug 257155 we should know about it immediately.
Ignoring the regression detection, I’ve always found unit tests to be fantastically useful when developing new code or fixing bugs. Zipwriter is a prime example, with a large number of tests that I can run by typing a single command I can test whether the changes I have made have solved the problem and not introduced any other errors.
The next step of course is to start adding unit tests for the extension manager. I have some already in progress and hopefully soon some of the key parts of the EM should be getting checked on a daily basis.
Since the disclosure of potential vulnerabilities in the way Firefox (and other Mozilla applications) automatically update your add-ons we have been discussing how to tighten up the system in a way that is hopefully unnoticeable to users and not much extra work for add-on authors.
After a process of listening to authors on the newsgroups, forums and by email we now have a rough proposal of what changes we are looking to make. There’s still a few minor details to be ironed out of course. This is mainly of interest to add-on authors since there is an impact depending on how you host your updates. I’ve started threads on the newsgroup and forums so if you want to discuss the proposal there then that’d be good. I’d prefer it if you didn’t edit the main page of the wiki but feel free to stick small comments onto the discussion page.
Spam is one of those evils of the modern age. It looks less and less likely that a real 100% effective solution will be found which is a little sad but not a major deal to my mind. I’ve managed to turn off my old junk email accounts and train my filter to clear out 90% of the junk I receive. Surprisingly I’ve had more problems with spam comments on this site than I have in my email lately. Even with comment moderation turned on the Tab Sidebar extension was receiving a silly amount of junk comments. It probably still is but I’m now using a simple blacklist to catch it all.
I tend to think of spam in two categories. You have your normal junk, you know offers to help someone transfer $1,000,000 (ONE MILLION US DOLLARS) out of Nigeria and to enlarge various of my body parts. Then you have the stuff which is from reputable companies which have generally got your email legitimately, maybe I filled in something to download some trial software or maybe I was even interested at one point. Perhaps surprisingly it’s the latter of these types that irritate me more . In particular when I attempt to unsubscribe and I still keep receiving mails.
Take this example. When I left my last job (about a month ago) I cancelled my small business mailings from Microsoft as they weren’t relevant to me anymore. What do I find in my inbox today:
Dear Mr Dave Townsend
As part of a routine data inspection we have noticed you have elected to stop receiving communications from Microsoft.
So umm let me get this straight, you recognise that I have asked to not receive any more mails from you and as such you have decided to email me?
As Microsoft launches exciting new solutions and initiatives, there’s no better time to register for information that will give your business a critical technological advantage. We invite you to consider receiving communications from us again.
Oh well if you are wanting me to start receiving mails again then of course it’s acceptable for you to mail to ask me.
if you still choose not to receive communications from Microsoft, simply do nothing and we will not contact you again
Oh well that’s ok. Of course I haven’t done anything since I opted out and you still mailed me.
Really large organisations should know better. If I tell you that I don’t want to hear from you again, what I mean is “I DON’T WANT TO HEAR FROM YOU AGAIN”.
Last night (or I should say early this morning) I got a chance to watch some of the webcasts from the Firefox Developer Conference in Tokyo. I had been wanting to attend one of the previous developer days but had to pull out due to accomodation problems so it was really great that webcasts were available, I believe this is the first developer day to do so.
I managed to catch most of the sessions on FUEL and XULRunner and some of the presentations on the development environments that extension authors use. The first two were good to watch, simple overviews of topics that can often get too bogged down in details. I won’t say I learned a great deal new from them, but then given my background I wouldn’t expect to. I really hope they make the recordings available so when the next person comes onto IRC asking what XULRunner is we can point them to it.
The presentations on development environments were somewhat more challenging to watch remotely, the webcast didn’t really have the resolution to bring out the detail of what the authors were doing and there was some time lagging on the video so what was being described didn’t correlate with what you could vaguely see on screen. Plus it was nearly 9am here so I had to call it a day at that point and get some sleep.
All in all what I saw of the event looked excellent. Top work by the translators, I don’t know if they were volunteers or hired in but they did a great job of handling all the technical terms that were flying about. I hope to make a dev day soon but till then lets keep up this theme of webcasting them so those who can’t be there still get a chance to be involved. One thought from the future, how about taking questions from the remote viewers over IRC?
We’re looking at the situation of automatic updates for add-ons and whether or not tightening up the security of such updates is a good idea or not (this is one good reason why it could be). After some initial talking I would like to get a little feedback from the add-on authors out there who host their add-ons on their own websites and not on addons.mozilla.org. Are you such a person? If so then please take a few moments to check out the thread in the forums or on the newsgroup and take a few moments to answer my questions.
Of course nothing is a foregone conclusion at this point and you might have noticed that I host my own extensions, not on addons.mozilla.org so I totally have the self-hoster’s needs at heart 🙂
I guess it goes without saying that I’m fairly technically literate and as such I’m pretty well versed in both what makes a strong password and actually using them. I actually have a pair of passwords, one that I use for what I consider my more important logins (company accounts, servers and the like), and another that is for lesser services that if I lost or it got hacked then it wouldn’t be a major compromise of anything.
Given this it’s always particularly disappointing when I find something that I really want to use a strong password for but can’t, because the service in question can’t handle how strong my password is.
Take my new bank account with Lloyds TSB. The password for the internet banking is 6-15 characters, must contains letters and numbers, but cannot contain any spaces or anything non-alphanumeric. Bang goes about 4 characters from my strong password.
Lloyds aren’t alone either. I also have a savings account with Citibank. To log in to their online banking I am not allowed to type in my password by hand, instead I must use an onscreen keyboard with my mouse. Now I’m not quite sure what this is meant to serve, all it does is enter the characters into a regular html input box, you know, easily readable from an add-on or other form of spyware. And even worse the keyboard gives me just 51 possible characters to choose. At least Lloyds let me use both upper and lower case!
Maybe all these places having quite different restrictions on what characters I can use in my password is a cunning ploy to make me use a different password everywhere, but I find it a little disturbing that I’m able to use a stronger password with my online pizza delivery place than with my bank accounts holding thousands of pounds of savings.
Wow, it’s been a month to the day since my last post here, and quite a lot’s happened in that time. Those of you that keep up on Mozilla things might realise that I have changed jobs and I’m now working for Mozilla on the Firefox team under Mike Connor. I’m going to be putting work into the addons side of Firefox 3, in particular taking some of the main requirements as well as tackling some of the really irritating issues that have lain dormant for a little too long for my liking. Most exciting stuff for me right now (yes I know, I’m sad!) is that I’ve been working on doing unit tests for the extension manager component which makes testing new patches far easier to my mind as well as of course allow us to start catching regressions.
Getting this new position has been quite a fantastic achievement in my eyes and it’s allowed me to do other things that I’ve been needing to do for some time, like move house and various other personal goals that I won’t bore you with here.
In case you were wondering how this affects my extensions, well not much in all honesty. They are still all my personal work and all done in my personal time and the amount of time I have spare to work on them is (unfortunately) still about the same. I am however thinking about a fundamental change about how my extensions are available to the general public and in particular one that I think will encourage more outside contribution to my extensions, meaning that the burden is taken off me as a lone developer to add features and fix the bugs. I’m still mulling this over at the moment so watch this space for further news.
Update: Mozilla now produce intel gecko SDKs so there is no need to use the version I have put here, I’ll leave it for posterity though.
It’s currently a bit of a pain building xpcom components in intel macs. The only officially available sdk is ppc only. Until Mozilla come up with an official version, here is an intel build of it for those that want it: gecko-sdk-mac-intel-22.214.171.124.zip
As the name suggests it’s built against Gecko 126.96.36.199. To the best of my knowledge it’s right but please don’t bug me if you can’t get your component to work with it unless you’re pretty positive that it’s the sdk that’s wrong.
Right now I have no clue how you’d go about making a universal sdk, maybe if you know of a simple way then you could get in touch.
I can’t believe it’s been over a month since I wrote something here. Well I kinda can, lots of hectic stuff going on at my work right now which has been making finding time for Mozilla stuff tricky. Hopefully not for much longer.
I’m glad to say that I have managed to make great progress on the zip writer component. I have decided that dealing with multiple platform compiles for Nightly Tester Tools is just a bad idea, so instead I have pushed on with submitting the zip writer to Mozilla for review. Hopefully that will make it into tree where I (and of course anyone else) can just use it. There’s been a bunch of changes between the version I posted earlier and that that’s gone up for review, not least of which is a set of testcases that have made sure I didn’t break the old by making some cleanups.
Lots of people have been bugging me about when new versions of my extensions are going to be done. Sadly I can’t really do this. The old adage of “It’ll be done when it’s done” certainly applies. For most I don’t have a good handle on how much work is left and I certainly don’t know how much time will have to spend on them in the near future which sort of messes up any planning.
Oh and I’d just like to say hello to all you people reading planet out there. Assuming I haven’t broken my atom feed you should be able to see everything I write here from now on, you lucky lucky people!