things that interest me this week – 29 oct 2014

Quick Update: A couple of people mentioned there’s no Mozilla items in here. They’re right, and it’s primarily because the original audience of this type of thing was Mozilla. I’ll make sure I add them where relevant, moving forward.

Every week I put together a bunch of news items I think are interesting to the people I work with, and that’s usually limited to a couple wiki pages a handful of people read. I figured I may as well put it in a couple other places, like here, and see if people are interested. Topics focus on the web, the technologies that power it, and the platforms that make use of it. I work for Mozilla, but these are my own opinions and takes on things.

I try to have three sections:

  • Something to Think About – Something I’m seeing a company doing that I think is important, why I think it’s important, and sometimes what I think should be done about it. Some weeks these won’t be around, because they tend to not show their faces much.
  • Worth a Read – Things I think are worth the time to read if you’re interested in the space as a whole. Limited to three items max, but usually two. If you don’t like what’s in here, tell me why.
  • Notes – Bits and bobs people may or may not be interested in, but that I think are significant, bear watching, or are of general interest.

I’ll throw these out every Wednesday, and standard disclaimers apply – this is what’s in my brain, and isn’t representative of what’s in anyone else’s brain, especially the folks I work with at Mozilla. I’ll also throw a mailing list together if there’s interest, and feedback is always welcome (your comment may get stuck in a spam-catcher, don’t worry, I’ll dig it out).

– k

Something to Think About

Lifehacker posted an article this morning around all the things you can do from within Chrome’s address bar. Firefox can do a number of the same things, but it’s interesting to see the continual improvements the Chrome team has made around search (and service) integration, and also the productivity hacks (like searching your Google drive without actually going there) that people come up with to make a feature more useful than it’s intended design.

Why I think people should care: Chrome’s modifications to the address bar aren’t ground-breaking, nor are they changes that came about overnight. They are a series of iterative changes to a core function that work well with Google’s external services, and focus on increasing utility which, not coincidentally, increases the value and stickiness of the Google experience as a whole. Continued improvements to existing features (and watching how people are riffing on those features) is a good thing, and is something to consider as part of our general product upkeep, particularly around the opportunity to do more with services (both ours, and others) that promote the open web as a platform.

Worth a Read

  • Benedict Evans updated his popular “Mobile Is Eating the World” presentation, and posits that mobile effectively ”is” everything technology today. I think it needs a “Now” at the end, because what he’s describing has happened before, and will happen again. Mobile is a little different currently, mainly because of the gigantic leaps in hardware for fewer dollars that continue to be made as well as carrier subsidies fueling 2-year upgrade cycles. Mobile itself is also not just phones, it’s things other than desktops and laptops that have a network connection. Everything connected is everything. He’s also put together a post on Tablets, PCs and Office that goes a little bit into technology cycles and how things like tablets are evolving to fill more than just media consumption needs, but the important piece he pushes in both places is the concept of network connected screens being the window to your stuff, and the platform under the screen being a commodity (e.g. processing power is improving on every platform to the point the hardware platform is mattering less) that is really simply the interface that better fits the task at hand.
  • Ars Technica has an overview of some of the more interesting changes in Lollipop which focus on unbundling apps and APIs to mitigate fragmentation risk, an enhanced setup process focusing on user experience, and the shift in the Nexus brand from a market-share builder to a premium offering.
  • Google’s Sundar Pichai was promoted last week in a move that solidifies Google’s movement towards a unified, backend-anchored, multi-screen experience. Pichai is a long time Google product person, and has been fronting the Android and Chrome OS (and a couple other related services) teams, and now takes on Google’s most important web properties as well, including Gmail, Search, AdSense, and the infrastructure that runs it. This gives those business units inside Google better alignment around company goals, and shows the confidence Google has in Pichai. Expect further alignment in Google’s unified experience movement through products like Lollipop, Chrome OS, Inbox and moving more Google Account data (and related experiences like notifications and Web Intents) into the cloud, where it doesn’t rely on a specific client and can be shared/used on any connected screen.


Deadsquid Update

No one will really care about this but me, but that’s ok, I need to get in the habit of using this thing again.

The dedicated server that’s continued on something like 14 years of hosting for friends and family is entering it’s final days. I’ve been moving sites over, and am hoping coop is paying attention to his email. I ended up going with a Canadian firm for a VPS provider, and the experience so far has been pretty painless.

The biggest issue I’ve run into was slow MySQL response time, and I spent most of yesterday reading up on tuning and tweaking my.cnf. I wish I had read all this stuff about 5 years ago, as a lot of lights were turned on with regards to perf on cthulhu; I could have saved myself a lot of pain by changing about 10 lines. Thankfully I’m never too old to learn.

All mail services on cthulhu have been migrated, and everything else gets shut down at the end of August. I think everyone’s ready. It’s been fun, but I won’t miss it, and the VPS will be about 20% of the cost of a dedicated box with performance that appears to be at least comparable.

If you’re using deadsquid for anything and haven’t heard from me regarding migration, you should drop me a line sooner rather than later. Two more weeks and everything’s gone.

Decelerating Deadsquid turns 14 this month, and there are some changes afoot. I’ve been running a bunch of ISP-ish services – everything from dialup access to mail and web hosting – since 1995, and am tired of it. I like having my own server to do my own thing, and have been happy to help friends and family out by hosting their content, but the services I offer aren’t comparable to what can be had commercially at a very (very!) reasonable cost. So, to that end, I’m shuttering the dedicated server everything runs off of by the end of April.

Thanks to everyone who has used over the years to vent about Ingenia, keep in touch with friends, play Expert CTF, check your mail, and post your thoughts and photos. I’ve enjoyed it, but I’m not paying anywhere near as much attention to it as I should, and it’s hard to justify the $1200-1500/year it costs to maintain. If you have a website still hosted here, I’ll be in touch shortly, and will help however I can to move it over to a provider who will take much better care of you.


revisiting search

Mozilla Firefox has had a search bar since its initial release, and has helped to change the way our users look up information by giving them a single interface to a variety of search services. It’s also had search services in the location bar, but they’re not as accessible or (arguably) useful as what’s offered by the search bar. There have been minor tweaks along the way to how these work, but nothing fundamental has changed with respect to search features in Firefox since its original release. We need to change that. Starting now.

Search behaviours have changed, and there are a number of new – and sometimes even different – search service providers out there. Unfortunately, our users don’t always realize how many options are available to them through Firefox, the websites they visit, and search-related add-ons. Our users today are using search to find other people, stuff that’s close to them, what everyone’s talking about right now, and a host of other things, and we should be making that as easy as possible for them.

Aside from the great UI work being considered, like moving the search bar into the location bar, I think there’s considerably more we need to do across a number of domains. I’d like for us to start exploring how we improve the use and utility of search in Mozilla’s products and services, particularly:

1. What kind of information are our users searching for, and who is best-positioned to provide the most relevant information for those searches?

The web has matured in the last five years, and people are using different search services for specific tasks. We should categorize the search services that are available in Firefox, and ensure they’re relevant to the task at hand for the people who use them. The Fennec team has developed its search interface with this in mind, providing search services for different tasks that our user base performs on a regular basis, and I think this is something we should build on.

2. Are we doing a good job meeting the needs of the users in each locale we support?

Every locale we support starts with the default list of providers we offer in the US English version of Firefox. Our amazing localization teams have created these lists to add search services that are more relevant to their locales and the users in them, and they do a great job. I’d like to ensure we all have a better understanding of who those providers are, and what, if any, alternatives there are per locale. From there we can build on the categorization process, and provide a truly global list of search services for our users.

3. How can we help our users discover and use the options available to them?

There are several search features in the browser, along with search options other than what we provide by default in our search bar. We need to make it easy for users to add to the list of search engines to the list of defaults we provide, to discover that there are add-ons that enhance search utility, and that they can change things like the default search provider(s). The mechanics behind these features could be improved considerably, and we should make changes to both to make them more usable by our publishers and users alike.

4. What does our search wish list look like?

We should think hard about what we’d like to change in Firefox to make search better, as well as where we should incorporate search services and which services should be offered. We don’t need to constrain ourselves to how we’ve done things thus far, and should consider including anything and everything that will help. If we could get things just by asking, would they include things like:

  • add-on searches and discovery
  • suggestions on error pages
  • better user control of search preferences
  • context-sensitive searches by website
  • searches from within new tabs
  • insert your idea here (and in the comments!)

The net result of this process should be a list of new services and features we can incorporate into the Mozilla project and its individual products, and would encompass all of the information assembled. The idea is to get people thinking, and come up with a public plan for improving search across the board to keep our products relevant and useful.

None of these ideas are new, and have been considered at different times by individual groups or people. They touch the user, the product, our content providers, localization and add-on communities, and almost every functional organization at Mozilla, and requires the input of same. As such, they’ll always be considered individually unless we shift from a tactical mode of thinking to something a little more strategic. I’d like to kick things off so we can start driving towards that.

The sky should be the limit, and we shouldn’t constrain ourselves to any particular mindset.

So, how do you think search should work?

xbox 360 video stuttering on the panasonic pt-ae3000

I have a new PT-AE3000 projector from Panasonic, and to date I have loved everything about it except for a few games (most notably rock band 2) using the Xbox 360 with a progressive signal. The video (but not the audio) would stutter/jump at times, which could throw my timing off. It wasn’t a huge deal, but it was annoying.

I had disabled the usual culprits, namely noise reduction and Panasonic’s “Frame Creation” (used to make moving images less after-imagy), but was still getting stutter. Searching various forums didn’t really help, so I went back to the manual, and eventually found the culprit.

The projector has a setting in the “Options” menu for “Frame Response”, which is described in the manual as “You can minimise the time delay of image displayed for the progressive signals.” This feature has two settings:

  • NORMAL Prioritise the image quality
  • FAST Prioritise the frame response

Setting it to “FAST” cleared up the problem, and I didn’t notice any discernible difference to image quality in-game.

PowerLinc 2412U PLM and Indigo/OSX

Just a really quick hit, because it took me a little while to find this post, which pointed me in the right direction.

If you’re trying to use the PowerLinc 2412U on OSX 10.5 with the latest beta (4.1) of the Indigo home-automation and control server, you’ll need to install USB serial drivers first. It would have been nice if this was a little more apparent with either the Indigo software or the PLM’s documentation, but it’s not (or, if it is, it’s buried).

Get the USB serial drivers from here, install them, and now you should be able to see the interface and the proper port in the preferences panel. Hopefully this will save you a bit of the aggravation I went through trying to figure out why the hell I couldn’t initialize the interface between Indigo and the PLM.

removing the norton internet security 2008 toolbar from firefox

I run Winbloze. I know, I know, but I have had a much better experience with Windows environments on my desktop than I have had with my Ti or my MBP. Windows comes with risks, so to help mitigate (again, I know, I know) those risks I use Norton Internet Security.

My subscription ran out a couple days ago, so I upgraded to NIS 2008. NIS 2008 includes anti-phishing protection, which I already have. NIS wants me to use its anti-phishing, so helpfully installs a toolbar in Firefox. It doesn’t tell me it’s going to do this, nor does it give me a choice, and there’s no way I can see to remove it from within NIS, only disable fishing protection. This pisses me off, so…


If this kind of behaviour pisses you off as well, and you want to stick with the built-in protection Firefox offers, you can remove the toolbar. Fire up your trusty file mangler, and head on over to the Firefox installation directory (usually C:\Program Files\Mozilla Firefox). Under that directory is a subdirectory called “components”. In the “components” directory is a file called “coFFPlgn.dll”; move it somewhere safe and restart Firefox.

Toolbar gone!

Thanks Symantec, next time try providing it as an uninstallable plugin or, even better, an extension. Fucktards.

Update: Toolbar back! Every time you reboot, it “helpfully” restores the file. I have asked Symantec how I turn it off permanently. I think I know the answer already.

Update 2: Johan has a good workaround, as follows: After playing about for a bit I tried an old trick, close all instances of firefox, go to the components folder as mentioned above, and remove the coFFPlgn.dll file. Create a new empty file, name it the very same, coFFPlgn.dll and set it to be read only.

the null file is ignored by ffox, and marking it read-only will prevent Norton from over-writing it. works nicely for me. recommended. thanks, Johan!

dspam 3.8.0 as content filter with postfix breaks quarantine re-training

Ugh. No matter how much I try and cover all the bases, some little thing always turns what should be a 2-3 hour transition period into a 12 hour ordeal. Everything went pretty smoothly until final testing on re-training false positives with dspam. I had tested it with the existing quarantine, which was created under dspam 3.6.x, and everything worked fine.

So I flipped the switch, and as mail started to filter in after the switchover I tested the training app and, well… telling dspam that a mail it had flagged as spam wasn’t actually spam didn’t work. It’s a long, sordid story, so I’ll cut right to the chase.

This problem only appears if you are using postfix as the MTA, and running dspam in daemon mode via LMTP using the configuration outlined in ./doc/postfix.txt

Here’s the symptoms:

  • Attempting to retrain false positives in the quarantine fails. You can select the message for retraining, but when you submit the form the page simply reloads with the message still in quarantine.
  • Attempting to retrain false postives from the history page are reported as successful (message is marked as retrained in the history), but the message remains in the quarantine list and is not delivered to the mailbox.
  • Attempting to view the message in quarantine results in a blank screen
  • ^M characters are in the user’s .mbox file (this is the root cause of the problem)

The problem with 3.8.0 is that it adds <CR><LF> as an EOL to the user’s .mbox file instead of just <CR>. This breaks how dspam.cgi parses things. John Peacock came up with a fix for dspam.cgi that enables it to parse the messages properly, but that addresses the symptom, not the root cause, which is dspam itself. If you’ve had any messages delivered to your users without the dspam fix, you’ll probably want to modify dspam.cgi with the John’s patch until it’s reasonable to think the users have dumped their quarantines (or care about the messages in them).

There is a patch for dspam that corrects the problem, but it is unfortunately not in CVS, nor in the release tarball. The patch was only posted to the dspam dev list, and was a little hard to find until John sent it to me. For reference, this information is available at the mail-archive for dspam-dev under “Patch for ^M in Quarantine / Blank Quarantines”. The patch file is also at the bottom of this post, and you’ll need to patch dspam.c in your source and recompile. This will correct the linefeeds being inserted into the quarantine, and everything will play nice moving forward.

Huge, huge thanks to John Peacock for providing the trail, and for answering mail on a holiday pointing me at the fix. I was kind of banging my head over this from 2am through 8am. Good times.

Patch after the jump: Continue reading

ipod ressurection

so. my poor little ipod has been sitting in my desk drawer for a year and a half because I thought it was borken. what with all the clicking-hard-drive sounds it made when you tried to turn it on, this was understandable. this weekend I went through my office on a purge mission, and said ipod’s fate was put on the table.

when it worked, I lurved it. it had 40 gigs of space to throw all kinds of crap I’d never listen to on it, and some crap that I would. this craptacular bliss lasted about a year and a half, at which point the relationship ended abruptly. it has been replaced by two nanos since (the first one I can only assume is still in use) which are better than I could have thought possible.

I was going to chuck the 4g, but figured I’d look at repair prices becaue, really, who wants to throw away $400 after 18 months of use? I could buy a new drive for $150, and the next question was whether I’d fix it myself (preferred, ’cause I’m geeky) or send it in for the repair (“included in the price of the drive” say the ads – remember this part). so… could I repair the damn thing? heck, could I open the damn thing?

using the handy post over at ipod lounge, I realised I had a guitar pick, and could probably use it in the manner described. I found said pick and, sure enough, opened my 4g without scratching or breaking a thing. when I picked the shell apart, I was surprised to discover that the connector was actually loose. I re-seated it, and figured “what the hell, let’s give it a go”. I put it back together, and plugged it into a USB port (the battery was long dead).

the apple appeared, and then… so did the music. it works flawlessly, and I deleted everything on it and re-transferred 25 gigs of music to make sure the drive wasn’t teasing me. still working flawlessly.

so I’m wondering, how many repairs where people send their ipods in are simply a cable re-seat, or yield the repair shop a perfectly good drive… it’s probably not a huge issue, but I can’t help but wonder if the business card trick is really necessary, and if it’s just disconnecting and re-seating the connector that makes things happy.

I’ll use the 4g as the car player for the next month, and let you know how/if it continues to work. regardless, colour me “pleasantly surprised”.