tag:blogger.com,1999:blog-30019643755509004762024-02-19T06:38:58.226-06:00Loving the PenguinAdventures in LinuxMithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.comBlogger139125tag:blogger.com,1999:blog-3001964375550900476.post-28252254282693024522019-05-04T11:25:00.000-05:002019-05-04T11:26:47.117-05:00Firefox extensions certificate expired quickfixThe <a href="https://blog.mozilla.org/addons/2019/05/04/update-regarding-add-ons-in-firefox/" target="_blank">proposed solution</a> to Firefox's <a href="https://bugzilla.mozilla.org/show_bug.cgi?id=1548973">certificate expiration issue</a> won't work on Debian because data reporting is disabled for Debian builds of Firefox. A workaround mentioned in passing <a href="https://news.ycombinator.com/item?id=19823701" target="_blank">here</a> appears to have worked for me.<br />
<br />
The workaround consists of visiting <a href="https://storage.googleapis.com/moz-fx-normandy-prod-addons/extensions/hotfix-update-xpi-intermediate%40mozilla.com-1.0.2-signed.xpi">https://storage.googleapis.com/moz-fx-normandy-prod-addons/extensions/hotfix-update-xpi-intermediate%40mozilla.com-1.0.2-signed.xpi</a> directly and allowing Firefox to install the xpi file.<br />
<br />
Drive carefully. Your mileage may vary<br />
<br /><div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-55347367008805729462018-09-11T18:22:00.001-05:002018-09-11T20:00:48.973-05:00Solving "device not managed" for Ethernet connections in Network ManagerAt some point or another, Network Manager in Debian sid stopped managing my Ethernet connection. The source of the issue is <code>/etc/NetworkManager/NetworkManager.conf</code>: <br />
<pre><code>[main]
plugins=ifupdown,keyfile
[ifupdown]
managed=false
</code></pre>
<br />
The <code>[ifupdown]</code> clause seems to be telling NM not to do it. The solution is to turn the frown upside down, but doing so in <code>NetworkManager.conf</code> means the setting will likely be clobbered the next time NM is updated. The solution to that is to create a file in in <code>conf.d</code> named <code>10-globally-managed-devices.conf</code>. NM upgrades should leave custom files in <code>conf.d</code> alone.<br />
<br />
<a href="https://ubuntuforums.org/showthread.php?t=2358631" target="_blank">Source</a><div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-51837031859099894582018-08-06T16:37:00.002-05:002018-08-08T11:25:01.673-05:00The new su PATH behaviorThe <code>su</code> command that's now being packaged with Debian sid has significant change in behavior. In the past, when you did an:
<br />
<pre><code>$ su</code></pre>
the PATH would get set to root's PATH. Now, it remains the <i>user's</i> path:
<br />
<pre><code>$ su
Password:
# echo $PATH
/home/mithat/bin:/usr/local/bin:/usr/bin:/bin:/usr/local/games:/usr/games:/snap/bin</code></pre>
This means you won't pick up <code>/usr/sbin</code> where a lot of commands you'd typically use with root privileges live.<br />
<br />
The fix to this is that <b>you now have to append a dash to the command</b>:<br />
<pre><code>$ su -
Password:
# echo $PATH
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin</code></pre>
I don't know <strike>whether this impacts <code>gksu</code> or</strike> if <code>sudo</code> and/or <code>gksudo</code> are affected. <code>gksu</code> <i>is</i> affected as well, but with <code>gksu</code> there doesn't appear to be a way to get back to the old behavior.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-70899856412484364152018-01-12T18:08:00.003-06:002018-08-03T14:34:25.412-05:00Fixing Skype for Linux scalingAfter a long hiatus from (actually a deliberate avoidance of) Skype, I need to install it again -- at least temporarily. And the latest version of Skype for Linux is appears to be an Electron app, which means its scaling on my non-standard 118 dpi screen is seriously wonky.<br />
<br />
The typical fix for this with Chrome/Chromium and Electron apps is to pass the <code>--force-device-scale-factor</code> option to the executable. Applying this option to Skype works as well, but you need to dig a bit to find where to add it. I ended up modifying <code>/usr/bin/skypeforlinux</code> so the last line reads:<br />
<br />
<code>nnohup "$SKYPE_PATH" --force-device-scale-factor --executed-from="$(pwd)" --pid=$$ "$@" > "$SKYPE_LOGS/skype-startup.log" 2>&1 &
</code><br />
<br /><div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-74056763620152382552017-01-23T14:15:00.000-06:002017-01-23T14:18:16.027-06:00Fixing Chromium remote extension loading in Debian sid<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgngs4ux1Ce-qkGTzNSmREmfZx8Kph5b7YHEE9BUCS5Ngemu8CYpPmUmLhYCmM1Lz1hlaVI0007jfAbBbfOVZzuDNBxxyewMM4wLZ3LeW6IfuwnSzJNkVXkHUJoi4SzIg9VCasGbAnDwaGe/s1600/chromium.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEgngs4ux1Ce-qkGTzNSmREmfZx8Kph5b7YHEE9BUCS5Ngemu8CYpPmUmLhYCmM1Lz1hlaVI0007jfAbBbfOVZzuDNBxxyewMM4wLZ3LeW6IfuwnSzJNkVXkHUJoi4SzIg9VCasGbAnDwaGe/s1600/chromium.png" /></a></div>
<br />
The newest Chromium in Debian sid disables remote extension loading by default. This has the effect of disabling extensions <i>en masse</i>. I'm not sure what the reasoning is, but it's damn inconvenient.<br />
<br />
It's claimed that one way to get the old behavior back is to add<br />
<br />
<code>--enable-remote-extensions</code><br />
<br />
to the list of flags in <code>/etc/chromium.d/default-flags</code>. This will affect all users on the system, and the file could be overwritten by an upgrade. In addition, until I know better what's behind the change, I'd really like to apply the change only to my account, not to all users.<br />
<br />
I did this by adding:
<br />
<pre><code># fix latest Debian Chromium disabling remote extensions
export CHROMIUM_FLAGS=$CHROMIUM_FLAGS" --enable-remote-extensions"</code></pre>
to my <code>~/.profile</code>. Be sure to log out and back in to experience the change.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com2tag:blogger.com,1999:blog-3001964375550900476.post-88934143777612217052016-10-25T21:51:00.002-05:002016-10-25T21:52:59.292-05:00Fixing PackageKit authentication in Debian sid under XfceToday after updating stuff, my <code>pkexec</code> commands (e.g., <code>synaptic-pkexec</code>) stopped working. I don't know whether this was caused by something I inadvertently did that borked things or whether it's a bug in Debian sid. I figured out that the root of the problem (no pun intended) is that a policy authentication agent was no longer loading on login -- in spite of <code>policykit-1-gnome</code> being installed and the entry for <i>PolicyKit Authentication agent</i> being checked in Xfce's Session and Startup autostarts.<br />
<br />
Reinstalling stuff didn't help.<br />
<br />
Further sleuthing revealed that the <i>PolicyKit Authentication agent</i> entry in Xfce's Session and Startup was pointing to a (now) nonexistent file: <code>/usr/lib/policykit-1-gnome/polkit-gnome-authentication-agent-1</code>. Further examination revealed that the file <code>/etc/xdg/autostart/polkit-gnome-authentication-agent-1.desktop</code> installed by <code>policykit-1-gnome</code> also was pointing to the same nonexistent file.<br />
<br />
So here's what I did to kludge a fix: <br />
<ul>
<li>Created a new directory <code>/usr/lib/policykit-1-gnome</code></li>
<li>Added a link from <code>/usr/lib/x86_64-linux-gnu/polkit-gnome-authentication-agent-1</code> to <code>/usr/lib/policykit-1-gnome/polkit-gnome-authentication-agent-1</code>.</li>
</ul>
After logging out and back in, things worked as expected.<br />
<br />
I could have added a custom entry into my Xfce autostarts that pointed to the correct file, but that wouldn't have addressed the second issue of <code>/etc/xdg/autostart/polkit-gnome-authentication-agent-1.desktop</code> pointing to phantoms.<br />
<br />
To undo the changes once/if the problem is better sorted, all I need to do is delete <code>/usr/lib/policykit-1-gnome</code>.<br />
<ul>
</ul>
<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-45941775915075917662016-05-28T18:40:00.004-05:002016-05-28T18:43:11.315-05:00Fixing virtual console resolution with nVidia driverI tried a number of things to get a laptop with an nVidia GT218M [NVS 3100M] video card to provide decent resolution in virtual consoles. This particular unit needs to run the proprietary video drivers to behave well--the big exception being the big-and-fuzzy virtual terminal text.<br />
<br />
What finally worked for me in <code>/etc/default/grub</code> was<br />
<br />
<pre><code># If you change this file, run 'update-grub' afterwards to update
# /boot/grub/grub.cfg.
# For full documentation of the options in this file, see:
# info -f grub -n 'Simple configuration'
GRUB_DEFAULT=saved
GRUB_SAVEDDEFAULT=true
GRUB_TIMEOUT=5
GRUB_DISTRIBUTOR=`lsb_release -i -s 2> /dev/null || echo Debian`
GRUB_CMDLINE_LINUX_DEFAULT="quiet"
GRUB_CMDLINE_LINUX=""
# Uncomment to enable BadRAM filtering, modify to suit your needs
# This works with Linux (no patch required) and with any kernel that obtains
# the memory map information from GRUB (GNU Mach, kernel of FreeBSD ...)
#GRUB_BADRAM="0x01234567,0xfefefefe,0x89abcdef,0xefefefef"
# Uncomment to disable graphical terminal (grub-pc only)
#GRUB_TERMINAL=console
# The resolution used on graphical terminal
# note that you can use only modes which your graphic card supports via VBE
# you can see them in real GRUB with the command `vbeinfo'
#GRUB_GFXMODE=640x480
GRUB_GFXMODE=1600x900
GRUB_GFXPAYLOAD_LINUX=keep
# Uncomment if you don't want GRUB to pass "root=UUID=xxx" parameter to Linux
#GRUB_DISABLE_LINUX_UUID=true
# Uncomment to disable generation of recovery mode menu entries
#GRUB_DISABLE_RECOVERY="true"
# Uncomment to get a beep at grub start
#GRUB_INIT_TUNE="480 440 1"
</code></pre>
The important parts are the <b><code>GRUB_GFXMODE=1600x900</code></b> and <b><code>GRUB_GFXPAYLOAD_LINUX=keep</code></b> lines.<br />
<br />
Don't forget to
<br />
<pre><code>$ update-grub</code></pre>
Credit goes to <a href="http://unix.stackexchange.com/questions/17027/how-to-set-the-resolution-in-text-consoles-troubleshoot-when-any-vga-fail">StackExchange</a>.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-16381852782123192762016-05-01T14:46:00.000-05:002016-05-02T13:13:56.744-05:00Adding "Run" options to Thunar Clicking on executable shell files <a href="https://bugzilla.xfce.org/show_bug.cgi?id=7596#c15">no longer executes</a> them in Thunar. <br />
<br />
While there is a <a href="http://docs.xfce.org/xfce/thunar/hidden-settings">hidden option</a> to revert to the old behavior, I decided in a new install to instead add right menu "Run" and "Run in terminal" custom actions. First create the custom actions from within Thunar to get valid <code>unique-id</code>s, and then open <code>~/.config/Thunar/uca.xml</code> and edit the entries as needed:<br />
<br />
<!-- HTML generated using hilite.me --><div style="background: #ffffff; overflow:auto;width:auto;border:solid gray;border-width:.1em .1em .1em .8em;padding:.2em .6em;"><pre style="margin: 0; line-height: 125%"><span style="color: #000080"><action></span>
<span style="color: #000080"><icon></span>system-run<span style="color: #000080"></icon></span>
<span style="color: #000080"><name></span>Run<span style="color: #000080"></name></span>
<span style="color: #000080"><unique-id></span>xxxxxxxxxxxxxxxx-x<span style="color: #000080"></unique-id></span>
<span style="color: #000080"><command></span>exec ./<span style="color: #800080">&apos;</span>%n<span style="color: #800080">&apos;</span><span style="color: #000080"></command></span>
<span style="color: #000080"><description></span>Execute this file<span style="color: #000080"></description></span>
<span style="color: #000080"><patterns></span>*<span style="color: #000080"></patterns></span>
<span style="color: #000080"><other-files/></span>
<span style="color: #000080"><text-files/></span>
<span style="color: #000080"></action></span>
<span style="color: #000080"><action></span>
<span style="color: #000080"><icon></span>xterm-color<span style="color: #000080"></icon></span>
<span style="color: #000080"><name></span>Run in terminal<span style="color: #000080"></name></span>
<span style="color: #000080"><unique-id></span>yyyyyyyyyyyyyyyy-y<span style="color: #000080"></unique-id></span>
<span style="color: #000080"><command></span>xterm -T <span style="color: #800080">&apos;</span>%n<span style="color: #800080">&apos;</span> -e <span style="color: #800080">&quot;</span>./<span style="color: #800080">&apos;</span>%n<span style="color: #800080">&apos;</span>; echo <span style="color: #800080">&apos;&apos;</span>; read -p <span style="color: #800080">&apos;</span>Press <span style="color: #800080">&lt;</span>Enter<span style="color: #800080">&gt;</span> to close...<span style="color: #800080">&apos;&quot;</span><span style="color: #000080"></command></span>
<span style="color: #000080"><description></span>Execute this file in a terminal<span style="color: #000080"></description></span>
<span style="color: #000080"><patterns></span>*<span style="color: #000080"></patterns></span>
<span style="color: #000080"><other-files/></span>
<span style="color: #000080"><text-files/></span>
<span style="color: #000080"></action></span>
</pre></div>
<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-87398290026221039792016-04-26T11:58:00.000-05:002016-04-26T15:39:29.542-05:00Concerns about Qt's increasing commercial focus<div class="separator" style="clear: both; text-align: center;">
<a href="http://www.publicdomainpictures.net/pictures/110000/velka/road-warning-sign.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="http://www.publicdomainpictures.net/pictures/110000/velka/road-warning-sign.jpg" height="213" width="320" /></a></div>
<br />
It's not just me. <a href="https://www.phoronix.com/scan.php?page=news_item&px=Qtio-Open-Source-Concerns">Phoronix</a> is reporting on growing concerns in the Qt community about the project's increasing commercial focus and apparent deprecation of its open source core.<br />
<br />
I have said many times in these pages and elsewhere that I really like Qt: it's the right tool for a lot of jobs. But increasingly I've been feeling that any contribution I make in educating users about Qt is benefiting a commercial enterprise and not a community tool. So much so that I am re-surveying other open source multi-platform libraries to use as an alternative to Qt.<br />
<br />
If the community no longer has meaningful Qt ownership, a lot of users will no longer be interested.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-74470416311559382922016-04-25T16:30:00.004-05:002016-04-27T10:58:31.796-05:00Fixing Debian sid touchpad with libinputA recent sid update brought in Xorg's <code>libinput</code> package, which appears to make the old way of configuring touchpads (including <code>synclient</code>) obsolete.<br />
<br />
One of the issues I had after the update was that my touchpad's tap-to-click stopped working. I fixed this using the info on this <a href="https://unix.stackexchange.com/questions/210144/synclient-does-not-find-synaptics-properties-despite-synaptics-touchpad-in-xinpu">Stack Exchange post</a>, namely I created a file<br />
<code>/usr/share/X11/xorg.conf.d/10-touchpad.conf</code> containing:<br />
<br />
<pre><code>Section "InputClass"
Identifier "MyTouchpad"
MatchIsTouchpad "on"
Driver "libinput"
Option "Tapping" "on"
EndSection</code></pre>
<br />
Note the different location from the one in the post. I also disabled the old <code>50-synaptics.conf</code>, <code>50-vmmouse.conf</code>, and <code>51-synaptics-aptosid.conf</code> files (the latter two present possibly only in aptosid).<br />
<br />
Apparently, Petter Hutterer has a <a href="http://who-t.blogspot.com/2015/01/xf86-input-libinput-compatibility-with.html?utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+Who-t+%28Who-T%29">post</a> that goes into detail regarding what you can do with <code>libinput</code>, but I've not yet had time to read it.<br />
<br />
<b>Update (later the same day)</b>: With the above changes, my touchpad still had behavioral problems. I'll return to this later, but for now I re-enabled <code>50-synaptics.conf</code>, <code>50-vmmouse.conf</code>, and <code>51-synaptics-aptosid.conf</code> and purged <code>xserver-xorg-input-libinput</code> (which removed <code>xserver-xorg-input-all</code>).<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-85571009285355977972016-04-10T12:40:00.001-05:002016-04-26T11:58:40.871-05:00Qt "Demerger"<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhC9ZagL-rRmysYQ0zl6_SBkc1gSFs8gZB6xpsR6UX4r_IBKc1xa5qk1T-fBax0r0d6fdAGlyrCg9aE4R1gfGd5OX4S6kKxkYjddHe0QL_3l0JG1BsisI6GKbeVijupKOC-C9S0zPrZ7imS/s1600/separate-paths-29941281454668Ush8.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="202" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhC9ZagL-rRmysYQ0zl6_SBkc1gSFs8gZB6xpsR6UX4r_IBKc1xa5qk1T-fBax0r0d6fdAGlyrCg9aE4R1gfGd5OX4S6kKxkYjddHe0QL_3l0JG1BsisI6GKbeVijupKOC-C9S0zPrZ7imS/s320/separate-paths-29941281454668Ush8.jpg" width="320" /></a></div>
<br />
<br />
<a href="https://www.phoronix.com/scan.php?page=news_item&px=Digia-Qt-Demerger">Phoronix</a> reports that <a href="http://digia.com/en/investors/demerger/">Qt will be "demerged"</a> from parent company Digia soon.<br />
<br />
It seems there has always been interesting dynamics around Qt, and this I suppose is the latest chapter. I'm hoping this is more of a good sign than a bad one.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-53396064221085412802015-10-12T14:34:00.001-05:002016-04-25T16:47:05.759-05:00Midori browser font renderingI was having issues with Midori not rendering fonts the way I thought it should in aptosid XFCE. I set a <code>default.css</code> in <code>~/.local/share/midori/styles/default.css</code>:<br />
<pre><code>* {
font-smooth:always !important;
-webkit-font-smoothing: subpixel-antialiased !important;
text-rendering: optimizeLegibility !important;
}
</code></pre>
but that didn't fix what appeared to be wonky antialiasing. So I had a look in my ~/.Xresources and lo!:<br />
<pre><code>Xft.dpi: 96
Xft.antialias: true
Xft.hinting: true
Xft.rgba: rgb
Xft.autohint: false
Xft.hintstyle: hintslight
Xft.lcdfilter: lcddefaul !!! Typo here !!! Should be "lcddefault".
</code></pre>
Things are much improved now. I'm not 100% certain that it's 100% right, but it's definitely mostly right.<br />
<br />
The takeaway for you, dear reader, is not to ignore your display configuration in ~/.Xresources.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-28837201222208699412015-10-03T16:26:00.001-05:002015-10-03T16:29:10.992-05:00Broken Netbeans C++ code completion in Debian sidWhen I use the <a href="http://lovingthepenguin.blogspot.com/2015/09/workaround-for-netbeans-crashing-in.html">workaround</a> I wrote about for Netbeans crashing on startup in Debian sid, I notice that some C++ code completion doesn't work as
expected. In several classes I've tried writing, code completion isn't
picking up stuff declared in the header file.<br />
<br />
When I use a copy of
Oracle Java downloaded just for NB (i.e.,
<code>netbeans_jdkhome="/home/me/opt/oracle-java/jre1.8"</code> in <code>/home/me/opt/netbeans/etc/netbeans.conf</code>), things work as expected.<br />
<br />
Whether
this is a different or related bug I can't say, but I thought it worth
mentioning in case anyone else is having the same problem.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-36187799709695560472015-09-16T14:38:00.000-05:002015-10-03T16:29:23.828-05:00Workaround for Netbeans crashing in Debian sidA short while ago, my local install of Netbeans 8.0.2 started refusing to start with OpenJDK on my <a href="http://aptosid.com/">aptosid</a> box. I got around the problem by downloading Oracle Java and pointing Netbeans at that via my <code>netbeans.conf</code>:<br />
<pre><code>
...
# netbeans_jdkhome="/usr/lib/jvm/default-java"
netbeans_jdkhome="/home/mithat/opt/oracle-java/jre1.8"
...</code></pre>
<br />
Today I found <a href="https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=798924">Debian bug #798924</a> that suggests a workaround that allows using OpenJDK. In <code>/etc/java-7-openjdk/accessibility.properties</code>, comment out the line:<br />
<br />
<code>assistive_technologies=org.GNOME.Accessibility.AtkWrapper</code><br />
<br />
So far so good. <div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-24941637253880259172015-08-21T13:30:00.001-05:002015-09-16T14:43:07.502-05:00Fixing Qt5 themes in XfceA lot of my Qt apps (e.g., qpdfview, VLC) recently stopped respecting my GTK theme. I did some checking, and it turns out they are all Qt5 apps. Google led me to the <a href="https://forum.manjaro.org/index.php?topic=14931.0">QT5 apps don't obey GTK theme settings</a> post at the Manjaro Linux forum, which has some interesting things to say (generally applicable to all distributions).<br />
<br />
What I eventually did was add:<br />
<br />
<code># Workaround for Qt5 app theming.<br />export QT_STYLE_OVERRIDE=gtk</code><br />
<br />
to the end of my my <span style="font-family: "Courier New",Courier,monospace;">.profile</span>, logout and log back in. Works now.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-22978514404414125792015-08-16T20:50:00.000-05:002015-08-16T23:46:41.587-05:00Looking for a new RSS news reader<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEh3nRJTZ8gZINppM6D_V9z_6aM91CxAC8mo6KfvVwS2yejFfmwx-qsZjo-mlNCc5shsGxUf026jkjDK0M033DAntfROVxRaM4Hbywnh3JBtFSoHygGTTJx9KRCq_nXywVGgr6UNWfC-NDdF/s1600/science-and-technology.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" height="213" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEh3nRJTZ8gZINppM6D_V9z_6aM91CxAC8mo6KfvVwS2yejFfmwx-qsZjo-mlNCc5shsGxUf026jkjDK0M033DAntfROVxRaM4Hbywnh3JBtFSoHygGTTJx9KRCq_nXywVGgr6UNWfC-NDdF/s320/science-and-technology.jpg" width="320" /></a></div>
<a href="https://tt-rss.org/">Tiny Tiny RSS</a> is a very decent if not glamorous self-hosted, FOSS news reader. A while back the developer decided to make the install and update process git-based. In addition, "shared hosting accounts,
windows and other alternative OSes, free tiers of PaaS services of any
kind, is not supported. Not supported in this case meaning: it may work
in your particular case but if you have problems you are on your own."<br />
<br />
I don't know with any certainty if the deprecation of shared hosting support has anything to do with some weirdness I'm experiencing on the shared hosting account where I have my TT-RSS installed. So, I am preparing for the worst.<br />
<br />
I want a self-hostable and FOSS solution if at all possible. The solution also needs to play well with mobile (Android), either through really good Web site design or a dedicated app. I also insist on keystroke article navigation (next and previous).<br />
<br />
I'll update this list as I try things.<br />
<br />
<h2>
FOSS</h2>
<b><a href="https://owncloud.org/">ownCloud</a></b> sports a news module. I'm taking it for a spin through a free as in beer provider. The desktop experience is fine--apart from the ordering of articles, which is <i>almost</i> chronological. The mobile Web experience is nearly tolerable. There is a dedicated <a href="https://github.com/owncloud/News-Android-App">News app</a>, the usability of which is pretty good, but it has some arcane sync setting configs to get it to sync with the "main" (i.e., Web) list of read articles. You'd think it would do that by default.<br />
<br />
<a href="http://freshrss.org/"><b>FreshRSS</b></a> looks promising, but I am having issues with the app's not marking read articles as read. I don't know if this is a problem with my provider's DB connection. (It's not the same provider as my TT-RSS install). Keyboard navigation is fine, and the mobile Web experience is workable. It's claimed that the <a href="https://f-droid.org/repository/browse/?fdfilter=rss&fdid=org.freshrss.easyrss">EasyRSS</a> app can connect to FreshRSS, but I have yet to try it. The desktop web interface also has what I feel is a significant usability issue: if you set articles to me marked as read on scroll, then when you scroll the page to select the category of feed you want to read, you manage to mark whatever articles you scrolled through at the same time as read.<br />
<br />
<h2>
Proprietary hosted</h2>
<a href="http://reader.aol.com/"><b><b>AOL Reader</b></b></a> comes as a bit of a shock: it's quite decent. Has the needed features and the mobile Web experience is tolerable. It has a 1000 feed limit.<b><b></b></b><br />
<b><b><br /></b></b>
<b><a href="http://feedly.com/"><b>Feedly</b></a></b> comes highly recommended. Its Google login isn't too heinous as far as permissions it's granted are concerned. The Twitter login however is. The desktop experience is generally good. (I haven't tried it on mobile.) The privacy policy is riddled with potential share points. The service itself serves up mixed https and http content.<br />
<br />
<b><a href="https://www.g2reader.com/">G2Reader</a></b> also comes highly recommended. But it lacks keyboard navigation, so poop.<br />
<br />
<b><a href="http://digg.com/reader">Digg Reader</a></b> looks promising, but I don't like that the only ways in are via Twitter ("Read Tweets from your timeline. See who you follow."), Google ("View your email address, View your basic profile info, Manage your data in Google Reader"), and Facebook (no thanks).<br />
<br />
<b><a href="http://feedreader.com/">Feedreader</a></b> doesn't seem to get a lot of ongoing love from its developers, with the most recent news being from 2015 and 2013. This makes me think it could go away without much notice.<br />
<br />
<a href="https://theoldreader.com/"><b>The Old Reader</b></a> has a 100 feed limit for free accounts.<br />
<br />
<b><a href="http://www.newsblur.com/">NewsBlur</a></b> has a 64 feed limit for free accounts.<br />
<br />
<h2>
Current thoughts</h2>
If I can't get either of the FOSS options to work to my satisfaction, I may end up using AOL Reader--subject to its not being subject to awful privacy and terms of use policies. <br />
<br /><div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-87710981352307076902015-04-26T13:45:00.001-05:002015-04-26T13:45:41.849-05:00Debian Jessie<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjff-xmEoMoNz4sO9lOoueu6pAmdKIU2v0Z3H3q1yeQO1kKjTp5c0xp8aJ6bPgZQRajJYH2i-LlL74vOtUcGFu7zSllnfr9hvpx7on4Lc68RQYAotc1EZaqN4IEdfIGRNltxbDBA7Vyd7zj/s1600/debian-logo-101.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEjff-xmEoMoNz4sO9lOoueu6pAmdKIU2v0Z3H3q1yeQO1kKjTp5c0xp8aJ6bPgZQRajJYH2i-LlL74vOtUcGFu7zSllnfr9hvpx7on4Lc68RQYAotc1EZaqN4IEdfIGRNltxbDBA7Vyd7zj/s1600/debian-logo-101.png" /></a></div>
<br />
<a href="https://www.debian.org/News/2015/20150426">Debian 8.0 Jessie</a> is out!<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-44511164446976389122015-04-04T17:49:00.001-05:002015-04-04T20:43:26.224-05:00KB3033929 - or the joys of dual booting<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEi3axXJw0Y7BriWTEXK5R8V1fZSk2El23UJ48sWOpN92czmNdSu3WeCfN3978V1WvDnDQfevFAHOuK0f0Ug05CGuOb-pt-ig6JrkH_11msD29OVPTw4QdRpsxVWRIJVaWBlYPNRoh5KY5xL/s1600/wellington-boots.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEi3axXJw0Y7BriWTEXK5R8V1fZSk2El23UJ48sWOpN92czmNdSu3WeCfN3978V1WvDnDQfevFAHOuK0f0Ug05CGuOb-pt-ig6JrkH_11msD29OVPTw4QdRpsxVWRIJVaWBlYPNRoh5KY5xL/s1600/wellington-boots.jpg" height="211" width="320" /></a></div>
<br />
I have been bitten by the bug described in Microsoft's <a href="http://support.microsoft.com/en-us/kb/3033929">KB3033929</a>. <br />
<br />
What I did to fix it:<br />
<ol>
<li>Dug around and found my Windows 7 install disc. Luckily I had one. Woe be to you if your computer came without one.</li>
<li>Started my computer with the Windows 7 install in the CD drive. Did the obvious things until I got to a dialog that gave me the option to Repair your computer--which I selected.</li>
<li>In the following window, I selected Command Prompt and ran <code>Bootrec.exe /FixMbr</code></li>
<li>That seemed too easy, so I also ran the option to check for and correct startup issues.</li>
<li>Removed the Windows 7 install disc media After the automatic reboot (into Windows--bye, bye GRUB), then manually rebooted.</li>
<li>Prompted windows to try (again!) to install the offending update.</li>
<li>Rebooted to confirm that the update took.</li>
<li>Reinstalled GRUB using <a href="http://www.supergrubdisk.org/rescatux/">Rescatux</a>.</li>
</ol>
So far, no drama. <br />
<ol>
</ol>
<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-24751523282494037002015-02-07T00:46:00.001-06:002015-02-07T01:12:24.935-06:00Ubuntu phone?<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="http://www.bbc.com/news/technology-31148661"><img alt="Ubuntu phone" border="0" src="http://news.bbcimg.co.uk/media/images/80806000/jpg/_80806928_14d7a3e6-f757-4222-b77b-54902b68a63d.jpg" height="218" title="" width="320" /></a></div>
<br />
It looks like an Ubuntu phone is going to see the light of day. While I really want to see this launch be a success, the description of the phone by <a href="http://www.bbc.com/news/technology-31148661">BBC News</a> has me less than blisteringly hopeful. And not because the offering is a low-end device. Rather, it's because the phone won't have the most valuable distinctive feature promised by the platform--conversion to desktop mode when hooked up to external devices--and it will place front and center a largely app-less <a href="https://developer.ubuntu.com/en/scopes/">Scopes UI</a>--which seems to try to be different mostly for the sake of creating some kind of distinction.<br />
<br />
I think history has shown that different for different sake doesn't work out well in crowded, high-tech consumer markets. Different in and of itself doesn't offer the user value. In fact, different in and of itself represents friction for the user, a friction that something of greater value must overcome--be it improved performance, better usability, increased visceral appeal (with longevity), or useful new features. In addition, the problems of being different are amplified when when it comes from an upstart that lacks the marketing muscle to boom-crash-firework users into thinking it's not what it really is.<br />
<br />
Yes, you need product distinction to succeed in a crowded market. But it has to be a distinction that represents value, not novelty. Inaugurating the platform with a product that's crippled in the way this one is may forever give the platform a bad reputation--which would suck.<br />
<br />
Then again, maybe Scopes solves a problem I didn't realize I had. Which would be pretty cool.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-791708628750283942015-01-21T21:44:00.001-06:002015-01-22T13:15:13.793-06:00Samba and mtp automatic mountpoints with gvfs<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEj7qmSVH6fYnIUrO8ozkAl9s-oTHJi2KEpQ9UG2UMXZlHpXEJfSxteyZXEJ5uOJ96BHd0CxvLx-erOvvGb_IWRFppvmE-Gt0j80vXkGNcAws0hNDDCVECOB3f6j98zwdjD2HjSyW9ck-1UA/s1600/mtp-auto-crop.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEj7qmSVH6fYnIUrO8ozkAl9s-oTHJi2KEpQ9UG2UMXZlHpXEJfSxteyZXEJ5uOJ96BHd0CxvLx-erOvvGb_IWRFppvmE-Gt0j80vXkGNcAws0hNDDCVECOB3f6j98zwdjD2HjSyW9ck-1UA/s1600/mtp-auto-crop.png" /></a></div>
<br />
Apparently, if you browse to a Samba share using <code>gvfs</code> (e.g., from Gnome's file manager, Thunar, etc.), it has a good chance of appearing as a mounted volume in <code>/run/user/{your-user-number}/gvfs/</code>.<br />
<br />
It seems to work for <code>mtp</code> devices as well. I hooked up my Android phone, and I got a <code>/run/user/1000/mtp:host...</code> directory. I tried a couple command line operations in there, and while it was slow as snails, it worked. Now I can write scripts that operate on data stored on my phone!<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-36054406911674707692014-12-17T18:02:00.002-06:002015-01-22T12:33:42.101-06:00On the need for an open source smartphone<br />
<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiRhBFNu-JzJhfXuzoA-ULQWkR34BVh-79bHnqQvTiF09p5La_fDOKFP52UKWLbQFw9216CCTFOfHaAxO17zjs0QAvQh0oJBDhy7twQ5ATLwICav5Appn5yjlc1gYTUUcyvpyM1kxXVhd2q/s1600/a-small-door.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiRhBFNu-JzJhfXuzoA-ULQWkR34BVh-79bHnqQvTiF09p5La_fDOKFP52UKWLbQFw9216CCTFOfHaAxO17zjs0QAvQh0oJBDhy7twQ5ATLwICav5Appn5yjlc1gYTUUcyvpyM1kxXVhd2q/s1600/a-small-door.jpg" height="320" width="212" /></a></div>
<br />
<a href="http://threatpost.com/manufacturers-backdoor-found-on-popular-chinese-android-smartphone/109929">Threatpost</a> is reporting on <a href="http://researchcenter.paloaltonetworks.com/2014/12/coolreaper-revealed-backdoor-coolpad-android-devices/">Palo Alto Networks' discovery</a> of a backdoor on an Android phone sold primarily in the Chinese and Taiwan markets that allows the vendor (and ostensibly anyone approved
by or who impersonates the vendor) to take over your phone.<br />
<br />
Whether there is any hyperbole in this or not, it's clear that what they describe could very easily be done. And it thus underscores why we need a fully open source smartphone platform that <i>puts the user in control</i>. A solution that's effectively open only for vendors or leaves the last mile closed (i.e., drivers) just isn't good enough.<br />
<br />
Once we have the software solved, we can move on to building hardware that's open in the critical areas. But let's start with the easier to solve software problem. With just the smallest help from manufacturers, we could have this problem solved yesterday.<br />
<br />
<span style="font-size: x-small;"><a href="http://threatpost.com/manufacturers-backdoor-found-on-popular-chinese-android-smartphone/109929">Threatpost</a> (via <a href="http://beta.slashdot.org/story/211211">Slashdot</a>)</span><div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-55271203023514093242014-12-10T16:17:00.000-06:002014-12-11T13:57:55.576-06:00Here to stay?<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhuxyjZtleA_b2-F7tzQO5TdU-t7w5r6k4z2JZh9pNw2MXmvAGkXtd7UcGJRjmo5jW1wNYH1mUIXt_x00obK13zEdqrgeGpuexqKdcC5bl5AJ6bPghONrKh5ktPDcbULqFesUBBPbG84y7F/s1600/here.png" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhuxyjZtleA_b2-F7tzQO5TdU-t7w5r6k4z2JZh9pNw2MXmvAGkXtd7UcGJRjmo5jW1wNYH1mUIXt_x00obK13zEdqrgeGpuexqKdcC5bl5AJ6bPghONrKh5ktPDcbULqFesUBBPbG84y7F/s1600/here.png" height="200" width="200" /></a></div>
<br />
As part of my effort to diversify my online data siloing, I've been looking for an alternative to Google Maps/Navigation on my phone. I've slogged through a lot of FOSS and proprietary offerings, but all have had various inadequacies that left me lusting for the latest from Google.<br />
<br />
Then today I learned that the part of <a href="http://company.nokia.com/en">Nokia</a> that wasn't sold to Microsoft released a beta of their <a href="https://www.here.com/?x=ep&map=45.165,-93.32,10,normal">Here</a> mapnav app for <a href="https://www.here.com/app/en/?L=23">Android</a>. So far it has been quite good. All the features I want, some that I didn't know I wanted but am happy to have, and a reasonably usable if less than stunningly attractive UI. I only hope their monetization plans don't gut the app after it comes out of beta.<br />
<br />
Helpful links:<br />
<ul>
<li><a href="https://legal.here.com/en/terms/serviceterms/us/">Service terms</a></li>
<li><a href="https://legal.here.com/en/privacy/policy/us">Privacy policy</a> </li>
<li><a href="http://help.here.com/en/privacy/#q6190">Account and privacy</a></li>
</ul>
<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-23037169235221569332014-11-14T17:19:00.002-06:002014-11-14T17:40:18.889-06:00WebDAVI tried a number of things to get a speedy and easy-to-use WebDAV setup and finally settled on using the setup described in <i>Kasun's Tech Blog</i>'s <a href="http://techiech.blogspot.com/2013/04/mounting-webdav-directory-in-linux.html">Mounting a WebDAV directory in Linux (Ubuntu)</a>.<br />
<br />
I deviated from the above by leaving <code>/etc/davfs2/davfs2.conf</code> as it is and instead edited the conf file at <code>~/.davfs2</code> that magically showed up. I also found a <code>secrets</code> file there where I added credentials. I suspect these were added after the first access I performed, that is, after I did a <code>$ mount .</code> in the mount directory. <div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-22575234307506128302014-11-07T14:49:00.001-06:002014-11-07T14:58:14.641-06:00crontab -r<div class="separator" style="clear: both; text-align: center;">
<a href="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhf9wQ-3yF1kZvnsin52MZIYaRmlFcUkpKW6G1_jOCYZskj1Cp9p0g7W9POH2CrjfIVL_Pry-qXNJPIpKAXf0N2cE4XDObVe2Cihr4xcqqJKIirFa5ZHwR7TKDTDiPJrilTt47AkwQtU6D4/s1600/broken-brown-bottle.jpg" imageanchor="1" style="margin-left: 1em; margin-right: 1em;"><img border="0" src="https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEhf9wQ-3yF1kZvnsin52MZIYaRmlFcUkpKW6G1_jOCYZskj1Cp9p0g7W9POH2CrjfIVL_Pry-qXNJPIpKAXf0N2cE4XDObVe2Cihr4xcqqJKIirFa5ZHwR7TKDTDiPJrilTt47AkwQtU6D4/s1600/broken-brown-bottle.jpg" height="250" width="320" /></a></div>
<br />
<b><a href="http://linux.die.net/man/1/crontab"><code>crontab -r</code></a></b> is sooo close to <b><a href="http://linux.die.net/man/1/crontab"><code>crontab -e</code></a></b> — exactly one adjacent key to be precise.<br />
<br />
And <code>crontab</code> files are stored in <code>/var/spool/cron/crontabs</code> rather than inside the (backed up) <code>/home/mithat</code> directory?<br />
<br />
Crap.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0tag:blogger.com,1999:blog-3001964375550900476.post-63578279407320965702014-10-21T13:24:00.001-05:002014-10-22T10:09:33.068-05:00New laptop timeAlas, I think I need to upgrade from my 5+ year old current laptop. I need a dual boot system (with Windows) for teaching, so this means I will need to confront UEFI madness and a whole new slew of hardware compatibility foo--and maybe even a distribution change to one that has solid support for UEFI installs.<br />
<br />
Fun times ahead.<div class="blogger-post-footer">Copyright (c) 2020 Mithat Konar</div>Mithathttp://www.blogger.com/profile/11589391741274626742noreply@blogger.com0