Did you know Calibre can turn an RSS feed into an eBook? I didnt! It turns out Calibre, tt-rss, and Wallabag make it possible to roll your own news that you can read right on your eReader! #selfhosting #indieweb #technology
I’ve mentioned this before, but I’ll mention it again: I’m a big fan of RSS. For the uninitiated, RSS is a way to subscribe to a feed of content from a website and consume it in a reader or other tool of your choice. And despite claims that it’s dying out, I still manage to have more content in my feed reader than I possibly have time to consume.
For a long time I used Feedly as my RSS reader of choice. But back in October I decided to switch to tt-rss, a self-hosted RSS feed reading service that works on both browsers and through a mobile app. Then, in a fit of boredom, I used some self-hosted home automation tools to incorporate email newsletters into my feed. Meanwhile, I also decided to stand up an instance of Wallabag, a self-hosted website bookmarking service.
But I ran across a problem: with all this content at my fingertips, I started to fall behind, particularly on all those long-form articles and newsletters I want to read.
And then I discovered Calibre’s news scraping features and a solution presented itself!Continue reading...
Social media algorithms care only that you're engaged. They exist to advertise. Everything else is a side-effect. RSS lets you ditch the machine and build your own feed from trusted sources. #indieweb #politics #technology
Quite a few years ago, for personal reasons, I decided to drop out of major social media platforms. This was just at the time when those platforms truly started to take over the world, so the whole thing more or less passed me by as I watched from the sidelines. As a result, it wasn’t until very recently that I came to appreciate just how much these platforms have become the primary way that people run across content online.
Of course, this really shouldn’t be surprising. Once upon a time, the internet was made up of an untold number of websites, big and small. And this posed a real problem of content discovery. Sure, we managed. We managed with search engines, and bookmarks, and web portals, and other ad hoc technologies. But it was a huge pain.
Today, this same kind of content discovery is done on social media platforms, with content pushed to the consumer by machine learning algorithms that optimize for “engagement”, which is a technical term for “time spent on the service”.
On its face this would seem like a good thing! After all, if you’re engaged, that must mean you’re delighted by what you see!
But the reality is a lot more complicated. Yes, certainly the things that delight us will keep us engaged. But so do the things that make us outraged, or offended, or jealous. And the algorithm can’t tell the difference. So whether you’re clicking on a link because you want to see a picture of a large cat in a small box, or you want to read an outrageous article about how the world is really flat, it’s all the same to the machine.
The result is an algorithmic filter bubble that often serves to misinform, usually while making us miserable.
On the other hand, those algorithms really do provide a useful function: They push interesting content to us so we don’t have to go and seek it out. The problem is, we have no control over how they function.
Well, as you can probably guess, I’m here to tell you that there is an alternative, and it’s a technology that’s almost as old as the web itself: RSS.Continue reading...
With the proliferation of IoT (aka “smart”) devices, the #Futurama episode “Mother’s Day”, where robot toasters and staplers revolt against humanity at the behest of a crazed dictator, is starting to feel eerily prescient…#technology
I’ve been rewatching “Halt and Catch Fire”, and what I forget is, in the background, folks like Chuck Peddle, creator of the 6502, were the giants on whose shoulders Jobs and Gates stood. Thank you, Chuck!#technology
Great piece on Brave. The Freeloading argument isn’t accurate, but the rest of it is spot on and highlights how Brave is regressive and disempowers content creators.#technology
For those unaware, a couple months back I decided to get in on the OLPC G1G1 program, which allowed we priviledged members of the first world two buy two XO units, one which would go to a school somewhere in the developing world, and one that we could keep. The goal of the program was to spur interest in their work while encouraging donations, and let me tell ya, it sure worked on me! As it also did for two of my co-worker. :)
Anyway, I’m happy to report that a couple days ago, my XO arrived! And it’s a fine thing to behold. Compact and cleverly designed, these tiny, kid-focused, ruggedized laptops are a marvel of technology. Darn cute, too.
Anyway, there’s a ton of other pictures and material one these little things, so rather than going on and on about what makes them so cool, I’ll just post a photo and be done with it:
BTW, did I mention I typed this up on the XO? Well, aside from the photo upload, which apparently doesn’t work in the XO browser.
Geez, three weeks since my last update, and… well, frankly, not a whole lot has happened. Odd how, when you become an adult, your life suddenly becomes a lot less interesting from day to day. However, it goes without saying that at least a couple noteworthy things have happened, otherwise, why the post, other than to regurgitate my varied and disconnected thoughts into the digital ether?
So, what news? Well, first off, a photo from the garden. A while back, I decided it was about time to get some various plants planted. This included the usual garden, a mix of the everyday peas, carrots, spinach, and some zucchini, with rogue dill fragrantly invading the empty spaces, a hill of potatoes, just a few to see how they do, some sunflower plants, and last but not least, berry bushes, specifically raspberry and saskatoon. Well, in the weeks following, much has been afoot. The sunflowers are at least six inches tall, the various garden vegetables are coming along quite nicely, but most importantly, my raspberry bushes have started bearing sweet, delicious fruit. And to prove it, I even have a photo!
Incidentally, these are just the most recent of the ripened berries. I’ve already enjoyed a couple early fruit, and let me tell ya.. they are delicious.
Meanwhile, on the technology front, I finally decided to take the plunge and by my very own laptop. To be honest, the fact I hadn’t done this already was just a little bit ridiculous, and after spending many hours hacking away on the company tanktop (big, hot, heavy, and ugly), I finally decided the time was right to just buy one.
So, what did I buy? Well, I could’ve gone for a regular ol’ consumer laptop, something fairly inexpensive and functional, but such gear tends to be a little heavier, have poorer battery life, and most importantly, be somewhat fragile. So I decided to splurge on some gear that I knew would be heavy-duty, light weight, and generally capable of withstanding the abuses I’m likely to put it through.
I bought a Lenovo (previously IBM) ThinkPad. The T61, to be precise, which is the latest model in the T-Series.
Now, I know, ThinkPads tend to be a bit pricey, but for that, I got:
- A titanium roll cage,
- Best-in-the-business keyboard,
- Good battery life, which is even better with the Ultrabay battery I purchased,
- Quite thin and light, at just shy of 5 lbs,
- Quiet as heck.
Basically, exactly what I was looking for.
As for specs, it’s a 2Ghz Core2Duo, 2GB of RAM, 120GB hard disk, 1440x900, 14” wide-screen display (I would’ve prefered 1400x1050 standard-def, but they can’t be had… apparently the LCD manufacturers are shoving wide-screen displays down laptop manufacturers’ throats), nVidia Quadro NVS 140M video chipset, 802.11abgn, and a DVD-RW. Plus the usual assortment of USB ports, VGA out, a pair of PCI-Express slots, the Ultrabay, modem and NIC, etc.
As for software, it came loaded with Windows Vista business (which, for the record, isn’t nearly as bad as people claim, though I turned off Aero and SuperFetch pretty quickly), with the usual assortment of goodies, such as Office, and to my surprise, SQL Express 2005. In addition, after shrinking my NTFS partition by 40GB, I threw Ubuntu Gusty in, which, after a rough start, has worked fairly well (although sound doesn’t work at all, and I can’t adjust the screen brightness within X, which is pretty irritating). I even got suspend-to-RAM working after a bit of fiddling with settings in xorg.conf. I must admit, as a desktop OS, Ubuntu performs quite admirably, and is easily the closest I’ve seen to a truly mass-accessible Linux distribution.
So, overall, an excellent and worthwhile purchase! Of course, that basically blows my toy budget for the rest of the year.
I lied. The bastards gave me a trial version of Office. Oh well, OpenOffice it is!
Why did I install you, StumbleUpon? WHY???
Okay, so a little background, StumbleUpon is this browser extension that adds a toolbar to your browser. If you hit “Stumble”, it’ll search for websites it thinks you might like. Then you rate them. It has categories, so you can select particular subject matter, and you can even post comments and read what other people have to say.
Well, as you can imagine, this is an immense time waster. I mean, it’s really bad. It’s like the Del.icio.us front page, except less work. And the stuff I’ve found? Well, here’s a few gems:
- Taylor Hall Planet Perplex
- Kid Creatures
- The above lead me to: The Monster Engine
- Gummi Bear Sculptures
- T-Shirt Stencil Tutorial
- Crazy Bathroom. StumbleUpon actually found a blog entry on this subject, and some co-workers tracked it down on Snopes.
And those are just the things I thought were really cool. I just wish I hadn’t found yet another way to procrastinate at work.
I just came across this. It has photos and virtual tours of various Asian temples and other buildings. Very very neat.
So, today, one of my “friends” at work was so kind as to post a Google code search, with my name as a search term, on our internal IRC server (thanks a lot, Jeremy… jerk!). For those not aware, Google now has a specialized search engine that allows one to search through publically available source code (the bits that comprise the blueprint for a piece of software). It’s pretty handy for many things (computing various code metrics, finding interesting code snippets, and so forth). But, as it happens, it’s also a great way to find code authored by specific people. And, in this case, that specific person was me.
Well, this got me thinking: Imagine you’re applying for a job. Further, suppose, in your younger, less experienced years, you made some of your work available online. Perhaps you contributed to some open source project. Or maybe you released something of your own. Well, your potential employer now has a very easy way to find these bits and bobs, and may very well choose to include them as part of your evaluation. Now, that’s fine if all you’ve ever made available online is top-quality code. But for hacks like me, this can be a problem.
Of course, since the advent of the search engine, an employer has always had the option of digging around on the Internet for information about prospective employees, which is why it’s important to be careful about what you post online. But, for those in the tech sector, Google code search means their past work can now be more easily tracked down and evaluated.
And in case you were curious, you can see what Google has to say about me here.
Yes, it’s true! The MythTV frontend works! But what about the RAM, you ask? Well, I decided to take the stick back to Best to get a refund/swap/something. It was at this point that I discovered that, surprise!, I can’t get a refund! Apparently it was a final sale or something, which I evidentally didn’t realize at the time. This is especially shitty since I’m willing to bet that the stick is simply incompatible with the board, for whatever reason. But, they’re testing it anyway… and if it turns out to be good, I’m either going to try to get it swapped for a DDR2-533 stick or a store credit. And worst case, I could probably sell it.
Meanwhile, I decided to head to Futureshop and buy a stick of DDR2-533 ($71 “open box”, even though it had never been opened). My thinking was that, if I get a working stick out of Best, I can always return the new stick to Futureshop. After all, they’ll give me a refund. And, surprise surprise, with the new memory, the EPIA board POSTs just fine. Shocker!
The bright side is I now have a working Living Room Frontend! It’s not yet perfect, of course. The video output isn’t perfectly scaled to the screen size (apparently the TV-out chipset isn’t fully supported under Linux, yet.. yay!). DVD playback is very jerky (although the CPU isn’t pegged, so something else is going on there). And there are a bunch of things I haven’t finished, such as getting the VFD working, or enabling suspend-to-RAM.
OTOH, TV playback, itself, is perfect, with no tearing or stuttering, and the IR receiver I picked up works beautifully (although I need to adjust the receiver position a bit to improve reception). So overall, I’m pretty happy with it.
Update: Well, I got the VFD working! It was pretty darn easy, too. Lircd, the software I’m using to receive IR signals from the remote, has a driver for the display device, so I just needed to install lcdproc, and voila!, it works!
So, you remember that dead EPIA board, right? Yeah, the one I was going to use in my Living Room Frontend as part of my MythTV project? Well, after shipping it Fedex some time last week, it finally arrived at Logic Supply yesterday morning, and underwent testing. And can you guess what happened? Oh yes, I bet you can! The board booted just fine for them! Oooh, surprise surprise.
But, how can that be, you ask? Didn’t it exhibit some odd behaviour, such as powering up without the power switch being hit? Well, according to the support guy at LS, the board comes with AC loss auto-restart enabled by default! What this means is that, if it notices the AC get connected, it will automatically boot itself (which is good for a system you want on all the time). This mislead me into believing something was going wrong, when in fact it wasn’t. This, coupled with the fact that the board simply won’t POST without RAM installed, lead me to believe the board was toast when it was, in all probability, the RAM the whole time.
Damnit I hate hardware hacking.
Anyway, the bright side of all this is that Andy, another co-worker/buddy of mine, was visiting Princeton, New Jersey (where our corporate head office is). So, on the return path, I had LS overnight the board to Princeton for $25, and then I had Andy bring it back across the border. Result? Three day turn-around on the cheap!
Unfortunately, now I have a problem. I need to test my RAM. However, I’m not yet aware of a DDR2-compatible box that I can utilize for the purpose. And until I can verify the memory, I can’t really move forward on the FE. Did I mention how much I hate hardware hacking?
Well, today I decided it was time to get the IR blaster working in MythTV. This is the device that controls our settop box, so that we can tune channels in the digital tier.
Now, I decided to purchase an IR blaster (and receiver) from the guy running irblaster.info, and I gotta say, I couldn’t be happier! The blaster works absolutely perfectly, and I haven’t seen it miss a tune yet. Setting it up was remarkably straightforward:
- Plug into serial port.
- Install lirc kernel module.
- Copy DCT2524 configuration into /etc/lircd.conf
- Install channel.pl from [http://www.iwamble.net/IRBlaster_Howto.txt this tutorial] (along with some tweaks to make it behave well with our DSTB).
- Instruct Myth to use the channel change script.
And voila! Works like a charm. Tune times are a bit longer, now, as you’ve gotta wait for the box to get the key clicks and then switch, but overall, it ain’t bad at all.
Of course, this is all just testing. Until the replacement EPIA board arrives, we’ll be stuck watching regular ol’ TV for a while, yet.
Well, things are a bit stalled on the MythTV project, now. The Fedex guy came and took away the dead motherboard that was originally destined for the Living Room Frontend, so now begins the great EPIA Return Saga (tm). If all goes well, I’ll have a new board in a few weeks, with minimal fees in the forms of duty or taxes. I remain skeptical.
On the bright side, the IR receiver and blaster units arrived, so I’ll probably play with those this weekend. They look very well made, so I’m hopeful that they’ll work as advertised. ‘course, I was also hopeful that my EPIA board would arrive fully functional…
Meanwhile, the backend continues to work well. I continue to record The Daily Show and The Colbert Report during their late night showings on CTV (aka, channel 2, aka, one of the few basic cable channels I can currently access with Myth), and it seems to be doing the job quite nicely. I’ve also been noodling around with different plugins on the frontend, such as MythMusic and MythVideo, and everything seems to work as advertised.
Thus far, the only glitch was a period where the frontend started to stutter and generate prebuffering errors, but that could be related to running over a lowly half-duplex, 10 Mbs connection.
Well, the wiring is finally finished! With the help of my friend and co-worker, Chris, it took all of 15 minutes to get the coat hanger fished up to the outlet. From there, it was a cakewalk: draw string back through, draw ethernet up from basement, draw some spare string down into basement (for later expansion, if need be), rejoice.
In addition, last night I got the MythTV subnet running and operating correctly. So, once I get a new EPIA board, installing the frontend should be relatively straightforward. Assuming it isn’t DOA as well…
Speaking of which, it looks like shipping it back should be… interesting. The main problem is in potential customs and taxes which could be levied on the board as it returns across the US border. Hopefully, I can fill out appropriate paperwork and have it marked as a returned item, but we shall see. Meanwhile, on the way back to Canada, I will almost certainly be charged taxes and duty on the replacement, but luckily there’s forms I can submit to get those fees refunded.
Meanwhile, it looks like that EPIA board will spit out component video! It appears to have a header on the motherboard for attaching the proper outputs. Unfortunately… I don’t have a bracket to attach to said header. Fortunately, VidaBox came to the rescue! They specialize it building MCE-based multimedia boxes, but they also sell a number of accessories, one of which is the very bracket I need. A few emails back and forth between their sales staff, and voila! A bracket is on it’s way. After paying $15 for the bracket and $10 for shipping… US. Not to mention taxes.
Did I mention that ordering stuff online isn’t always the most economical thing in the world to do?
Well, after all the problems with ethernet cabling and bad motherboards, things took a bit of an upswing today on the MythTV project, and it all started when the TV tuner card arrived! Yup, it showed up before lunch today, and when we got home this evening, I promptly installed it in the backend and had it configured in around 15 minutes. It went beautifully! And the MythTV setup process went equally smoothly!
But it gets better! What I really wanted to do was test out the backend. So I plugged it in to our basic cable and then configured the mythtv backend. Then, I compiled the frontend on frodo (twice… I compiled 0.20 first, not realizing the backend was running the 0.20-fixes branch), and voila! I was suddenly watching TV on my computer! I could pause, rewind, skip forward, browse around in the EPG (which has a nice little preview of the current channel, just like our existing DSTB), and of course record. And it all works perfectly! Even the channel tune times, which I feared would be a little long, are decent… maybe 1.5-2 seconds to switch? Not bad at all!
So I’ve already marked The Daily Show and The Colbert Report to record this evening (since they run on CTV). We’ll see what they look like tomorrow. Then I can play around with the commercial skip and transcoding functions. Good times! Now if I can only get that EPIA replacement, I can be doing all this right on my TV!
I also managed to play around with MythWeb, the web interface to MythTV, and I gotta say, it’s pretty sweet. It provides a really nice interface for perusing your channel lineup, editting your recording schedule, viewing previously recorded material (assuming your browser and OS are set up correctly), and even accessing your music archive. Very nice! And, again, it worked more or less out-of-the-box (minus a probably unnecessary tweak to Apache’s configuration), proving once again that going with Fedora Core and pre-built binary packages was, hands down, one of the best ways to go.
Things were going so well. The MythTV Backend is now built and humming away quietly in my basement, Fedora Core installed and working nicely. The only minor glitch being some issues with the onboard NIC, though nothing that can’t be solved. Really, it was all going too well.
And then the other shoe dropped.
A couple of days ago, the board for my Living Room Frontend finally arrived after much waiting, whining, complaining, etc. When it finally showed up, I eagerly went home and mounted the board in the lovely Antec case I bought (at which point I realized the EPIA board could also fit in micro-ITX case… it looked so tiny in the mini-ITX Antec). I then wired up all the connectors, routed all the wiring nice and cleanly, and then went to install the RAM. Which didn’t fit. Why? Well, you see, I ordered DDR memory. I then decided to opt for the EPIA EN12000EG instead of the M6000. The M6000 takes DDR. The EN… takes DDR2. $80 blown. Doh.
So, today, after running some errands, we stopped by BEST and I picked up a stick of DDR2. Then, after dinner, I installed the stick and powered up the board. And nothing.
Actually, that’s not true. The PSU and case ventilation fans spun up, even though the power switch hadn’t been pressed. Not good. Experienced computer builders will immediately recognize the potential problem this presents.
So, I decided to start trouble shooting. The first thing was to make sure the PSU wasn’t at fault, so I disconnected the ATX connector from the motherboard and flipped the power switch. Nothing. Nada. This ruled out the PSU.
I then proceeded to reconnect the ATX connector and begin disconnecting other things gradually, testing the PSU in between. And every time, the fans spun up. Eventually, I was left with just the ATX connector attached to the board and nothing else. No RAM. No connectors. Nothing. And when I hit the power switch… the fans spun up. Conclusion? Bad motherboard. Grrr…
So now I have to return the board and get a replacement. Looks like no PVR for at least a few more weeks. On the bright side, at least I can get the backend finished up.
Incidentally, in the process of looking up resources on how to debug this problem, I found this forum post (second one down) describing the process of troubleshooting a motherboard. I mirrored the content here just in case the forum link disappears.
Meanwhile, I figured it would be a good idea to get the networking wired up to the living room. Now, my plan was to reuse the existing coaxial outlet as the ethernet jack. This is particularly convenient, in my case, because I’m dropping down between floors. You see, in this case, when dropping cable from scratch, it’s necessary to drive a hole between the lower wall framing plate and the subfloor. This means augering a hole through four inches of wood… not fun, especially if you don’t want to damage the wall. However, because I was reusing the coax connection, this hole had already been cut, making my job much easier.
Thus, all I needed to do was drop a piece of cat5 around twelve inches straight down into the basement. Easy, right? Well, unfortunately, it wasn’t that easy. You see:
- The coax utility box can’t be moved out of the way, because it’s fixed to the stud,
- I’m dropping through an exterior wall, which means insulation, which gets in the way,
- Because it’s an external wall, the box is surrounded by a PVC boot, making it more difficult to access from below,
- The hole in the basement is located near the exterior wall, over the existing framing, making it awkward to reach,
- The existing coax is fixed inside the wall, meaning it can’t be moved (or used to drag the cat5 through).
Now, the only workable method was to use a coat hanger to fish upward from the basement to the coax box. Once I reached it, the plan was to fix the cat5 to the coat hanger and draw it down into the basement. Things did not work out so well. I eventually gave up at 12:30 last night, after around 4 hours poking and prodding inside my walls.
So, what now? Well, Chris, a buddy from work, said he might come by on Saturday and give me a hand. Hopefully, between the two of us, we can get the cable run. Of course, until the EPIA replacement arrives, it won’t actually get connected to anything…
That’s right… things have started arriving! In particular, most of my NCIX-ordered items arrived, which allowed me to build my MythTV Backend and begin the software installation process. I’ve now got Fedora Core 5 installed and a bunch of services configured (though, without the Living Room Frontend ready, I can only get so much done).
Speaking of the frontend, the case has arrived! So, of course, I had to take a picture:
It’s bigger than I’d imagined, for some reason… about the size of a stereo receiver, and weighing in at a whopping 17lbs! Of course, this is, in part, because of the steel used in most of it’s construction. But, I gotta say… it’s pretty awesome. :) And it’ll fit perfectly in the TV stand.
Anyway, the EPIA board will likely arrive tomorrow (they apparently tried to deliver today, but I owe them money, presumably duty), and that only leaves the remaining NCIX items (which got shipped today) and the IR blaster/receiver, which are in the mail as I type this. At which point the fun will really begin!
Well, with all the hardware for the MythTV project on the way (mostly, anyway… NCIX didn’t have a couple items in stock. Like the TV capture card), it seemed like a good time to get the house in order for the new arrivals. Thus, as a logical first step, I decided it was about time I got my firewall rebuilt and moved all the various networking bits downstairs onto some kind of shelving. The result is this:
As you can see, I re-purposed an old Ikea shelving unit as a make-shift rack, and then moved all these things downstairs:
- Cable Modem
- 10Mbs hub (until I get a proper switch for the main house LAN)
- 10/100Mbs switch for the MythTV LAN
In case you’re wondering, this gear is all set up like so:
[[fig:Home Network Diagram]]
Why so complicated? Well, the primary complication is in my choice to shunt the MythTV stuff onto it’s own subnet. I had a couple reasons for this:
- The MythTV frontends will be configured via DHCP and bootp. Putting this on a separate network prevents conflicts with the main firewall DHCP (and any other network-booted devices I may deploy).
- I figured a dedicated, switched 100Mbs network for the Myth stuff wasn’t a bad thing.
Of course, if I had a proper switch which supported VLANs, I could have done this with a single switch, but I wasn’t prepared to pay the bucks for such a device.
As for the firewall, I ended up abandoning LEAF in favour of m0n0wall, a FreeBSD-based firewall that boots nicely off a mini-CD and writes it’s configuration to a dirt cheap USB flash drive I plugged into the back of the thing (IOW, no more unreliable floppies!). And as a bonus, it has a nice, easy to use web-based admin interface, so no longer do I have to hook a damn monitor up to the thing every time I want to reconfigure things.
Okay, this one’s a little different, though. I swear. It costs a lot more than my previous projects…
So what is it? Well, for many months now, I’ve talked on and off about building a DIY PVR setup. Usually this starts off with me complaining about the timeslots for The Daily Show and The Colbert Report, and quickly turns into a discussion of how awesome a PVR would be for timeshifting my favorite shows, listening to all the music I’ve transferred onto my computer, and watching all the content I’ve downlo… err… ahem. Anyway, as a result, I’ve often thought about:
- Buying and modding an XBox,
- Building a MythTV-based system, or
- Buying a Shaw PVR.
But it wasn’t until I got my pay adjustment, and the attendant backpay (all the way back to last december!) that it became a practical possibility. The problem is, now flush with cash, I think I’ve gone a little overboard.
It all started with plans to build a simple MythTV-based PVR with a mini-itx board of some kind, mounted in a little case. It didn’t take long before this ballooned into a full frontend/backend networked multimedia infrastructure project, with all it’s associated costs. But you know what? I don’t care, damnit! :)
Anyway, I’ve started some new pages to cover my MythTV project. There you can read about my motivations, details about MythTV, and the hardware I’m using. As for the current status, the gear has all been ordered and should arrive in the next few weeks, at which point, the fun begins! Meanwhile, I’ll need to string some cat5e between my living room and where the backend will live, so that should keep me busy.
And just to give you a little taste, here is a shot of the case I’m going to use for the Living Room Frontend:
Well, it finally happened. It was only a matter of time, really. Yes, that’s right… my floppy died.
First off, for those less geeky types, I should probably explain what I was using my floppy for. You see, hooking a computer directly up to the Internet is not unlike having unprotected sex with every woman in a two block radius. Why? Because all the computers in a two block radius are likely directly connected to yours (assuming you’re using cable internet), and so you’re vulnerable to any viruses, spyware, zombie computers, etc, etc, that happen to be buzzing around your local node. And I haven’t even covered non-local attacks.
Thus, it’s generally a good idea to use some kind of protection. This protection usually comes in the form of a firewall, which is not unlike a digital condom, acting as a layer of protection between your soft, vulnerable computer, and the harsh outside world. Now, there are two major kinds of firewalls. The first is a software firewall, and resides on the computer to be protected. Another is a separate firewall appliance which is physically located in the network path between the computer to be protected and the outside world. This would be this style that I favour.
So what about that floppy? Well, you see, as a geek, I thought it would be fun to build my own firewall. So I coupled some old spare parts with the Linux-based LEAF firewall package, and voila! Home-built firewall. And to improve protection (while, as it turns out, reducing reliability), I placed the actual firewall software on a, yup, you guess it, (read-only) floppy disk. Which has since died. :(
Fortunately, my wireless router can perform double duty as a simple firewall, so for now, this is my solution… though, at some point, I’d like to go back to a standalone firewall solution. Though, this time, I think I’ll put it on a CD-ROM.
I’m writing this entry prematurely, mainly because I’d already written the one for today, and this story needs to be told. At least, IMHO.
Okay, so, first some background. Yesterday our DVD remote mysteriously stopped working (well, not mysteriously… it had been slowly failing for a while). I tried the obvious and replaced the batteries, but it made no difference. Conclusion: remote is fux0red. Then, today, for some reason, the cable box remote seems to no longer be working properly. WTF?? Lenore then points out that this seems to coincide with my network noodling (as earlier blogged), and so I start to get a little paranoid. Did I wire something wrong? Is there IR interference being generated?? Because, if so, that means heat source, and heat source equals bad.
So, I begin the investigation. First, I test the remote at various angles. Slowly, I discover that it will only work when I’m standing in a certain position. So I move my body, but maintain the remote position. Doesn’t work. Yup, definitely interference. To verify, I shift the position of the cable box, and then fire the remote straight at it. Works perfectly.
At this point, rather than doing the smart thing and further investigating, I decide to begin testing solutions. I disconnect my networking job. No help. I disconnect the telephone line. Still no help. I disconnect the laptop cable. I unplug my Palm recharger. Nothing. Now I start to get more paranoid. Is it the wiring in the walls?? In the hopes that it’s not, I start closing blinds on the main floor. Maybe external interference?? It seems unlikely, but you never know… but, still nothing.
Now things get desperate. I need to narrow down the interference (this after 20 or 30 minutes of frantic confusion)! So, I position my body further from the TV, and find the position where the remote starts working. Then I step further back, repeat. Then I duck down while pressing buttons on the remote. Stops working. Stand up. Starts working. Duck. Works. Stand doesn’t work. What the heck?!? I’m definitely blocking something! So I look behind me… what could it be?!? I check through the kitchen, but there’s nothing obvious there, either. I start pondering cutting holes in the walls.
And then, I glance at the dining room table, and I notice something seemingly innocuous: the busted DVD remote control is on the table. Facing the TV. With brand new batteries in it. So I decide to turn the remote around. And sure enough, the cable box remote works perfectly. At this point, I yanked the batteries out of the DVD remote in a rage and slammed them down on the table… and what’s Lenore doing? Giggling. Giggling! While I’m trying to save our house from burning down. Well, damnit, the next time you’re suffering from mysterious IR interference, you can just stuff it!
So, aside from the computer building debacle as reported earlier, I finally got around to one other major TODO I’ve had on my list some time, that being to get some household networking going and move the damn cable modem and firewall downstairs. Previously, the cable modem was in our bedroom, and we had a hacked up piece of cat5 running into the den, which was a substandard solution, to say the least, so I felt it was about damn time to do something about this.
The beauty of this situation is that in basically all new houses, they’re wiring up the telephones using cat5, which means 8 pairs of wires, rather than just the old two. This means that, at every telephone jack in a new house, there are two pairs in use, and six extra pairs just sitting there, begging to be wired up. Well, regular ol’ 10baseT, which is capable of doing 10 Mb/s (sufficient for my needs) only needs four pairs. So, using the telephone line already wired into the den, I was able to hook up 10baseT from the den straight to the basement without having to drop a single line. Sweet!
For those wondering how to do this, it’s simple. You need just a few pieces of equipment:
- A modular faceplate and two connectors, an RJ11 and an RJ45 (or two RJ45s, if you like).
- A blade screwdriver.
- A pair of wire cutters/strippers.
With these items, the process of wiring is a simple matter:
- Remove the old plate and disconnect the wires. The blue and blue/white wires should be the ones in use. Warning: the ring voltage on telephone lines is enough to give a nasty shock, so do your best to avoid touching both wires at the same time.
- Connect the original wires to the middle pins of the to-be-telephone connector. For an RJ45, that would be as follows:
- blue -> pin 4
- white/blue -> pin 5.
- Plug in the telephone and verify it works.
- Wire up the ethernet connector as follows:
- white/orange -> pin 1
- orange -> pin 2
- white/green -> pin 3
- green -> pin 6
That’s it! Well, not quite. Now you get to wire up the other end. If you head to the electrical panel in your basement, you should see the various telephone lines from the house congregate. It’s up to you to figure out which one corresponds to the jack you’re wiring. I just disconnected them until I disabled the phone line I was working on. :) Once you’ve found the line, take the unconnected wires (there should be six) and splice a piece of ethernet to the white/orange, orange, white/green, and green lines such that the wire colours match. This will create a straight-through connection that you can wire into a hub. If you want to create a cross-over (so that you can connect the panel end directly into a computer) wire the white/green to white/orange and green to orange.
There, that’s it! After this, I installed a cable splitter, moved my cable modem and firewall into the basement, and then ran a patch cable from the jack upstairs into my hub, and voila! Done! Good times…
For my next trick, I think I’ll pick up another hub at some point, put it in the basement, and then put in another modular jack where my cable is currently wired in and run ethernet to the hub, in preparation for some sort of video PC or hacked Xbox-type solution. Plus, hey, it’s good ol’ techy fun!
So, in an amusing twist of fate, after going to Best Buy (yes, again… piss off!) and buying myself another video card for a whopping $129, I discovered (during a household search for other components, but I’ll get into that later) that I did, in fact, have a spare video card that will suffice… an old PCI Mach64, which will certainly do the job for the short term (and will eventually find itself in my server, in a swap for the GeForce card it currently possesses). Thus, now I find myself needing to go back to Best Buy (yes, for a third time) to return the card I just purchased.
Oh well, on the bright side, I’m saving myself $129, and I found that old card I was sure I had!
Bah, so I plugged in the Mach64 card, and the board wouldn’t power up. Odd, I thought. I pulled the card, and when I hit the power switch, at least the fan started spinning. So I plugged in the GeForce and… spinny fan, but no POST (Power On Self Test, for those not in the know… the part of the boot sequence where the RAM is counted, etc). Not even a beep from the speaker. And the HD led stays on, which doesn’t seem like a good thing.
So, I think I’m gonna abort this whole process. I’ll try taking the board back to BEST, and the card back to Best Buy (I’ll keep the RAM and just load up Frodo for now). Now, on to trying to compile a new kernel for Frodo, since the current kernel apparently doesn’t recognize more than 896 megs of RAM (as oppose to the 1.5 gigs that’s in there). I hate computers.
Further Updated Update:
Got the new memory in and the new kernel compiled. After futzing with my video drivers, I even have X working again! Now comes the wait to see if anything broke… good thing I kept the old kernel around.
So some of you may remember that a while back, I had a combination hard drive and power supply failure, simultaneously. The hard drive failure was pretty easy to detect, thanks to that lovely, disturbing clicking noise that haunts the dreams of anyone who’s experienced such a failure. Fortunately, the danger here was mitigated by the fact that, for some time now, I’ve chosen to run a pair of drives in a mirrored configuration (aka, RAID-1). Thus, while it appears to the user that I have a single drive, in reality, the data is always written to both drives.
The power supply, on the other hand, was an entirely different matter. When I noticed the failed drive, I removed it from the mirror and attempted to reboot my computer. But the other drive wouldn’t spin up! Or, it would spin up, but the computer wouldn’t detect it! Scared, I moved the drive to a spare machine I had, but sure enough, that machine wouldn’t detect the drive either! As a last resort, I took the drive to work the next day, and, to my great relief, the drive was perfectly readable, with all data intact. It was at this point that Lenore reminded me that my spare machine wasn’t in use because the hard drive controller was hosed. I then made the assumption that the same was true for my main computer.
Thus, I resolved to purchase myself another motherboard. So I took a trip over to BEST Computers and picked up a new board and a pair of new drives to replace my old mirror. But, when I got home that evening, I had a little epiphany, and decided to use my spare computer’s power supply in my main machine, just to test it out. And voila! It worked perfectly! Let this be a lesson: power supply failures can create weird, mysterious problems.
Anyway, what does this have to do with building computers? Well, suddenly, I had myself a spare motherboard and nothing to do with it. The natural thing, I thought, was to build a new machine (as opposed to just returning it…). So, eventually, I picked up a new power supply ($80), and this combined with the surviving hard drive from my last mirror, and the video card and RAM from my spare machine equalled a new box. Or so I thought.
So I began assembly. All seemed to go well. I got the motherboard mounted, and proceeded to grab the RAM… which, I discovered was 133-pin SDRAM, too old for my new board which required 184-pin DDR-RAM. sigh So I took a last minute trip to Best Buy (yeah yeah, piss off) and picked up a gig of new memory ($140 - $26 rebate).
Alright, so, RAM now installed. Case back panel, mounted. Front panel connectors, connected. Hard drive and CD-ROM, installed. So far so good. Lastly, video card.
Now, you probably already know this, but the job of the tech industry is to make simple things hard and hard things impossible. In the case of video cards, they decided to invent the AGP slot, into which a video card is to be installed. Which would be simple. To make it hard, they decided to have different voltages for AGP. 3.3v, 1.5v, and if that wasn’t enough, 0.8v too! So, if you have a card in one voltage, and board which only takes another, you’re hosed.
I bet you can guess what happened. I, apparently, have a 3.3v AGP card. Conveniently, my motherboard only takes a 1.5v AGP card. grumble. So now I’m stuck buying a new video card ($80).
So, total for this adventure:
Motherboard $150 Power supply $80 RAM $140 Video card $80 Rebate -$26 Total: $424
And the sick thing is, for about $80 more, I could get a whole new computer with a bigger hard drive and a nice sized LCD flat panel monitor to replace the 15” piece of CRT crap that I have now. And that is why I hate building computers.
So, in my on-going search for ways to justify the purchase of my PDA, I’ve decided to try and read my first e-book on the thing, specifically “The Da Vinci Code”.
Okay, quit laughing. I’m entitled to read a little pulp from time to time, too, ya know. So piss off! And, hey, it can’t be as bad as Decipher. No, seriously, it really can’t. If an author tried to write a book worse than that, I’m pretty sure his/her own lower intestine would reach up and strangle him/her, Douglas-Adams-style.
Anyway, surprisingly enough, the experience has been remarkably positive. I absolutely love real, physical books as much as the next guy (actually, probably more… I have this really nasty habit of “stopping in” to book stores and walking out with two or three new items to add to my collection. Which would be fine if paperbacks still cost $5, rather than the current going rate which is upwards of $10… frickin’ wallet rapists), and still think that the classic paper book provides a superior overall reading experience, although that’s probably at least in part due to nostalgia. But I have to admit, this whole e-book thing might not be so crazy after all.
Now, going in, I knew that e-books have some problems:
- Eye fatigue.
- Difficult to see in bright-light conditions.
- Poorer “random access” facilities.
- Less durable (for obvious reason).
In my case, the first two were my major concerns, especially given my poorer vision. But, as it turns out, it’s not as bad as I thought. The screen on my TX is clear and readable. The brightness controls make it pretty usable in a variety of light levels (though reflection is an issue). And as for the other issues, well, I can deal with them. Plus, e-books have a few advantages:
- Smaller pocket-print, thus easier to carry around.
- Easy to read one-handed, or even no-handed with autoscroll.
- Ability to adjust fonts, colours, etc, to suit the reader.
- Can carry around a whole collection of books easily.
- Very easy to just power on and read. The reader automatically remembers where I was and opens directly to where I left off.
- Works in dark environs. I could read while Lenore’s sleeping, if I wanted.
- Allows me to easily read material from resources like Project Gutenberg without having to print stuff off.
As for actual software, I really can’t say enough good things about PalmFiction. Unfortunately, the only things the author can say are in Russian, so you kinda have to fumble a bit with it. But once you do, wow! The feature set is incredible!
- Reads txt, PalmDoc, Word files, RTF, and others, and can even read compressed files.
- Can display the text using anti-aliased fonts converted from TTF sources.
- Supports any screen orientation, so you can read left- or right-handed.
- Can display in true full screen on hi-res devices. No wasted screen space!
- Does a great job of word wrapping and hyphenation.
- It’s FREE.
And there’s probably many more features I neglected to mention. Truely an awesome program, and far better than trying to read PDFs using PalmPDF.
On a separate but related note, the next e-book on my list is a Russian work called The Twelve Chairs, by Ilf and Petrov. ‘course, I was originally planning to read The Golden Calf by the same authors, as recommended by Arkadi, our resident Syberian. However, the folks at that site haven’t completed the translation, and the last thing I want is to be left hanging halfway through. ;)
The number of unprotected WAPs never ceases to amaze me. Here I am, blogging from a diner in North Vancouver on someone’s unprotected AP (yes, I know, questionable behaviour), and, frankly, loving it! If this keeps up, ‘net access will be ubiquitous in no time!