The 2019 Trackball Revival

Trackball and Mouse

In an attempt to head off some elbow tendon pain I started looking into a different mouse for the office. The suspect mouse is the Dell laser model that came with my work machine a few years ago. At home I use a much more ergonomic Logitech G400 that doesn't seem to give me issues. The Dell has very extreme angle on the front so it's a very unnatural fit for my hand. Maybe it works for people with smaller hands? It doesn't help that the DPI settings seem to be messed up and it will only swap between turtle in molasses and rabbit on meth. It's pretty hard to use in either of those modes. While I'm guilty of keyboard snobbery I'm generally OK with whatever mouse I can find that is comfortable. The Microsoft D66 and Logitech MX518/G400 are usually my go to as they fit my hand well, are relatively inexpensive, don't require any third party software and are basically everywhere. I guess it's time that I started getting into other relatively obscure input devices.

I haven't used a trackball since the late 90s. Back then I didn't care for the trackball as scroll wheels were becoming big and more and more pieces of software were utilizing them. At the time there really weren't any trackballs on the market with a scroll wheel. Optical mice were also starting to come on the scene and were popular among gaming enthusiasts. At the time trackballs didn't have scroll wheels which was a real sore point in early FPS games. We had a Logitech Marble back then, which is still available new BTW. It's a fine track ball but lacking the scroll wheel was a let down back then.

Now a days scrolls wheels are available on the few trackballs still on the market in the US (they're apparently still big in Japan and that's not a joke). There aren't many available as they've fallen out of favor with most people. Logitech makes a couple and there are some sellers online that sell Japanese import Elecoms. I went with a Logitech M570 as the MX Ergo was quite expensive. Plus I'm generally not a fan of non-user replaceable batteries and the rubberized coating the MX Ergo had. While I had used a Logitech Marble years ago I'd never used a thumb ball type trackball period so this was a first. So far I can say the M570 is a pleasure to use. It's only been a little over a week but I'm already comfortable with it. As far as gaming goes I've only managed to try a little Team Fortress 2 with it and I can see a trackball being a huge advantage in FPS games once you adjust. To be honest I'm not sure why these things are more popular than they are with gamers. The M570 is a tad small in the width department so my fingers tend to fall off the right hand side of the device but otherwise it fits my had well. I really don't care for the wireless part, especially since it's not Bluetooth and requires a small USB receiver, but it at least runs on a single AA that's user replaceable. Wireless trackballs don't make a lot of sense if you ask me. They don't move and aren't something you're going to use across the room, but it's what the kids like so whatever.

I think the key to preventing RSI is to stop the repetitive part of it so I'm not ditching the mouse completely. Changing position and devices will help in that direction quite a bit. I'm may be on eBay looking at some older models of trackballs too as most have been relegated to the dust bin of history, although some models can be quite pricey as they have a bit of a cult following. Elecom gets high marks on the new market but they can be pricey as they have to be imported. Really, if you want to one out I'd say give the M570 or a Trackman Marble a shake. You won't be out but for about $20-25 or so and if you like it there are other higher-end options out there. As an added bonus ne'er-do-wells won't really be able to mess you machine and I've already baffled a few people when I had to work in an open office setting this week ...

Unexpected Snow Day

Today was one of those days where a dusting to 1" turned into a bit more snow for Boone. All of these photos were taken through out the day with the Fuji x100s.

View from Summit

Theater Construction

East Hall from Sanford

Dog on a Walk

Ran into some folks out and about doing or wearing interesting things. This dog was pretty excited to get a move on.

Hat

Winter clothing can sometimes be interesting.

Snow Stack

As far as I could tell this is like that thing where people stack stuff on their cat but with a person.

2018 Holiday Random Print Mail Bombardment

Like getting random stuff in the mail? This year I'm going to mail out some prints to people who want them. They'll be either 4" x 6" or 5" x 7" depending on the paper I have laying around at the time. The image will be mostly random, could be pertinent to how we know each other or just whatever the heck I feel like printing. All of them will be photos I have taken.

If you want to participate please send the address of your mail hole or receptacle to leander@one-button.org.

In person delivery is an option if you are with in my laziness radius.

Images

Japanese Input on Fedora KDE

I've been taking a Japanese class over the last semester so I've had a need for Japanese input on laptops. I've been running Fedora KDE on my portable machines lately and I had no trouble finding documentation on changing up input methods on Fedora GNOME but there was little out there on the KDE spin. Some of the Kubuntu documentation got me in the right direction though.

The method I found for kata and kanji input uses fcitx, which have packages in the Fedora repos:

sudo dnf install fcitx-kkc kcm-fcitx

After those have installed add the Fcitx utility to the Autostart items in Plasma 5. Also, add the fcitx input modifiers in /etc/profile.d/fcitx.sh:

export XMODIFIERS="@im-fcitx"
export QT_IM_MODULE=fcitx
export GTK_IM_MODULE=fcitx

Restart the machine and open up the Fcitx Input Method configuration, you'll want to add the Japanese Kana Kanji as a secondary input method. If you don't see it come up in a search uncheck the "Only Show Current Language" option.

Input Method Selection

Now you should be able to hit Crtl+Space in any text input field and replace Romanji with Kana or Kanji characters. Additional presses of the space bar will cylce through different characters for the same sound. So you can have にほんご or 日本語 for example. Pressing Crtl + Space exists the Kana Kanji input mode. Pressing return after a typing a word in kana or kanji finishes the slection and allows you to move on to the next word. Very useful and no swapping keyboard layouts!

Input Method Selection

Autumn 2018

Just a few photos from the last couple of month around Boone, NC. All taken with either the x100s or X-T2.

Road Cones

Doggo in Coat

Blue Sky

Autumn Gradient

Stairs

Go Vote

Leaves on Ground

Early Snow

IBM Model M 1393464 Restore

This was originally posted some time ago on Deskthority. Hosting it here for posterity's sake.

Recently I stumbled on a 1393464 Model M off eBay for a decent price. This was a model with custom keycaps made for the American Airlines reservation system. Otherwise it's a normal 2nd generation Model M from 1990. I just think the differently printed keycaps look neat.

It arrived and I found that some of the keys were non-functional and I tracked it back to a dead line in the membrane. Since I'd never fully deconstructed a Model M before I decided to give it a go. Usually I just do the screw mod when the rivets break. Ordered the $10 membrane from Unicomp and it arrived Saturday. Just tonight I go the board all the way back together and it works! Typing this post on it right now. Below are some photos are comments on the process. It was more challenging than I thought it would be, getting the springs and flappers to stay in place while you re-attach the membrane and back plate is kind of tricky ...

1393464 Pre-disassembly

1393464 Pre-disassembly

Model information, this particular unit was built on Feb 06 1990

1393464 Pre-disassembly

Beige/cream barrel plate, this is a new one to me. I've only ever seen black ones ...

1393464 Pre-disassembly

Just a quick note on the new membranes from Unicomp: if you have an older M you're resotoring you'll need to trim off the rightmost four lines from the smaller ribbon cable. These membranes work with the older Ms but have the built-in lines for the LEDs on the new models. I just used a razor blade and some patience.

1393464 Pre-disassembly

New membrane pre-install...

1393464 Pre-disassembly

Keycaps back in and ready to be tested!

1393464 Pre-disassembly

All back together and ready to go. All in all not too bad of an experience and I found some threads on here helpful. Just thought I'd share the journey. Not sure how rare or special this particular Model M as it just showed up in my running eBay search. Found some info on the Clicky Keyboards site but that's about it. I think the membrane went bad due to some liquid exposure as it looks like there was some dried liquid in between the sheets. Anyway, should be good for another 28 years now I reckon.

Winter

Just a couple of shots from last week's snow storm.

Fujifilm X100s processed in Darktable. First real snow of the 2017-2018 winter at a few inches here in downtown Boone, NC.

Winter 2018 1

Winter 2018 2

RTL-SDR NOAA Weather Radio Streamer in Linux

If you’re into amateur radio you’ve probably heard of these cheap DVB-T tuner dongles re-purposed as software defined radios. They’re very popular for building scanners and streaming setups. I’ve got a couple of models that I use with Gqrx for listening to traffic on the local repeaters, weather radio and a few other things.

This weekend I finally had enough time to sit down and setup rtl_fm and try out a streaming solution for listening to NOAA weather broadcasts. I’ve streamed scanner and weather radio traffic using Icecast before but that was with a external radios and a mic input. This time I wanted to use a machine with no sound card (my home server machine). All of this was done on Debian Linux 8, to get started we need three pieces of software:

rtl-sdr ezstream icecast2

All are available in the default repositories so just install them with apt.

Next I had to configure icecast2, the most basic configuration should work but at least change your admin and source password in the authentication block:

 <authentication>
        <!-- Sources log in with username 'source' -->
        <source-password>password123</source-password>
        <!-- Relays log in username 'relay' -->
        <relay-password>password123</relay-password>

        <!-- Admin logs in with the username given below -->
        <admin-user>admin</admin-user>
        <admin-password>password123</admin-password>
   </authentication>

That should at least allow you to connect and get up and going. There are other options to secure and/or tweak but I’m not going to cover those here.

I used my NooElec Nano SDR as a test source. The rtl-sdr package comes with a program to handle FM tuning called rtl_fm. There are a few options to tinker with here but only too options are critical for operation here:

rtl_fm -f 162.500m -M fm -
-f 162.500m: sets the tuner frequency to 162.500Mhz 
-M fm: tells the tuner to use standard narrow FM tuning, if you want to listen to commercial radio you’d use wbfm or wideband FM. 
-: directs output to stdin

Send that output to lame to encoding:

lame -r -s 24 -m m -b 64 --cbr - - 
-r: assume raw pcm input 
-s 24: set the sample rate to 24K 
-m m: mono mode, NOAA doesn’t broadcast in stereo 
-b 64: set bitrate to 64kbps, it’s more than enough for this 
--cbr: constant bitrate 
- -: stdin/stdout

A quick note here, you’re going to have to mess with the sample rate in LAME to get things sounding right most likely. I arrived at 24K, much higher or lower and the pitch is off. It may not work with your model SDR, etc.

Lastly ezstream needs configuring. It took a while to get a working MP3 configuration sorted out but I eventually arrived here:

<ezstream>
    <url>http://localhost:8000/WNG588</url>
    <sourcepassword>password123</sourcepassword>
    <format>MP3</format>
    <filename>stdin</filename>
    <!--
      Important:
      For streaming from standard input, the default for continuous streaming
      is bad. Set <stream_once /> to 1 here to prevent ezstream from spinning
      endlessly when the input stream stops:
     -->
    <stream_once>1</stream_once>
    <!--
      The following settings are used to describe your stream to the server.
      It's up to you to make sure that the bitrate/quality/samplerate/channels
      information matches up with your input stream files.
     -->
    <svrinfoname>WNG from Mt Jefferson, NC</svrinfoname>
    <svrinfourl>https://hubble.buttonhost.net</svrinfourl>
    <svrinfogenre>Public Information</svrinfogenre>
    <svrinfodescription>NOAA Weather Radio from Mt Jefferson NC</svrinfodescription>
    <svrinfobitrate>64</svrinfobitrate>
    <svrinfochannels>1</svrinfochannels>
    <svrinfosamplerate>44100</svrinfosamplerate>
    <!-- Allow the server to advertise the stream on a public YP directory: -->
    <svrinfopublic>0</svrinfopublic>
</ezstream>

Save this to /etc/ezstream using your favorite text editor and pass it to eztream thusly:

ezstream -c /etc/ezstream.xml

The whole thing piped together looks like this:

rtl_fm -f 162.500m -M fm - | lame -r -s 24 -m m -b 64 --cbr - - | ezstream -c /etc/ezstream.xml

I just stuck that whole string into a shell script. If you want it to start at boot time you can shove it into /etc/rc.local for a quick and dirty solution.

Once all that is done it’s a simple matter of navigating to your Icecast server at http://whatever_url_u_have.com:8000 and clicking the m3u icon by the stream listed there. Open that file in whatever music player you want and enjoy. I use VLC, Rhythmbox or iTunes (when I find myself on a Mac) myself. Otherwise you can just check the weather app on your phone like a normal person. Next up I want to work on getting frequency scanning working so I can get the scanner back online.

Oh and you can check out the fruits of my work here: http://hubble.buttonhost.net:8000/WNG588.m3u

VLC WX

Windows 10 is pushy

Windows 10 is pushy and aggravating. So glad I only really have it in a VM for a couple of things. Thanks multi-core CPUs and VirtualBox!

"Would you like to make Firefox your default browser?"

-> Clicks yes

-> Doesn't actually make Firefox your default browser, but sends you to a control panel to do so.

Me: "OK, I'll just change it here" Clicks button to change it to Firefox

Windows 10 pops up an alert: "Whoa, wait a a minute there partner. Did you know Edge is the best thing since sliced bread? What sort of intellectual deficiencies do you have that makes you want to use something else? Are you sure that you really want to do this? I hear Hitler used Firefox."

Me: "Wut, OK this is getting silly. Just do it already!"

That took about five more steps than it should have. Linux and OS X (errr macOS) have their moments and problems but I don't know why Windows users put up with stuff like this. That last alert was obviously exaggerated for humor but it did indeed plead with me not to change away from Edge.

A Brief History of Things that Haven't Been ET

Position of HD 164595 in the constellation Hercules

With apologies to Dr Stephen Hawking for the title.

The media is running amuck with stories of an ET signal from HD 164595. It's an intriguing concept as HD 164595 is rather close to our sun on the main sequence with at least one known planet. However this outburst is unlikely to be from another civilization for a number of reasons.

First, let's look at few more promising ET candidates that turned out to be more regular natural phenomena. The most famous is probably Percival Lowell's martian canals. In the early 20th century photography was still in it's infancy so many astronomers still did visual observing. This combined with the human ability to insert patterns where none exists led Lowell to publish his findings as a civilization building canals. Not to fault him entirely as Mars does actually have a dynamic and constantly moving landscape but it's due to seasonal dust storms, not martians irrigating their crops.

The next promising candidate was the discovery of pulsars in the 60s. The first pulsar was found to pulse at 1.33 second intervals in the radio part of the spectrum. Antony Hewish and Jocelyn Burnell ruled out human made interference and jokingly named the radio source LGM-1 for "little green men." They didn't actually think the radio signal was from some far off alien intelligence, the name was just a joke. However popular media at the time certainly ran with that idea. Later, after more of these regular pulsing radio sources were discovered it was surmised they were rapidly rotating neutron stars. You know, just the corpses of long dead massive stars. Nothing terribly exciting there at all. Sarcasm heavily implied here in case you didn't pick up on that.

One of the more recent "it might be aliens" discoveries were so called perytons. They were only detected by the radio telescope at Parkes and only there since 1998. All sorts of theories were tested and after other radio telescoped failed to detect the signal and astronomical sources were excluded the team started to focus in on local phenomena. It turned out to be the observatory's microwave oven, I kid you not.

So with the colorful history of discoveries that might have been ET let's look at the HD 164595 signal. It's a strong microwave spike. So strong infact some have theorized that it could only be produced by a civilization that could harness the entire output of their parent star. It's rather unlikely that something capable of such engineering would still be using microwave transmission for communication. The signal also also isn't spectrally narrow like one would suspect an intentional transmission to be. There's also the fact that it doesn't particularly look like something you'd expect intelligent beings to be sending out. No obvious patterns. The argument could be made for it to be encrypted or encoded but if you're trying to get a neighboring star system's attention you probably want them to be able to understand it. Sure, it could have not been intended for intercept but why else would ET's be blasting out huge microwave bursts? An alien intelligence might use some mathematically significant signal or something else obviously artificially generated as a beacon to stand out against the noise. Random spectrally wide microwave bursts don't really do that. In fact, I'd say pulsars would have been a better candidate for SETI than this signal back in the day as their signals seemed to tick more boxes on the "might be ET" checklist. Sure, HD 164595 could be another intelligent life form trying to communicate with us but that requires extraordinary evidence. The signal could just as easily be Carl microwaving his burrito again.

RATAN_2

The suspect signal from the paper

Simplest explanation is that it's some natural phenomena we don't fully understand. We have observed exactly one star up close and personally and have in depth experience with one solar system. There is a lot we don't quite understand yet. This isn't some smoking gun for aliens as the media is screaming about. It certainly warrants further study though.

Page 1 / 5 >>>