Note: This is mainly for website updates and the occasional link. Most
of my personal updates are done on my Google+ page.
------==========/ Dylan Leigh - www.dylanleigh.net \==========------
Note: This is mainly for website updates and the occasional link. Most
of my personal updates are done on my Google+ page:
Someone just pointed out to me the top Google hit for "Yallara" isn't
the Wiki page on the animal, or the server itself:
On the other hand, Yallara is going to be killed in a month or so,
which is the opposite of awesome. :(
HOWTO: Number appendix pages with the appendix letter (A1, A2, B1, C1,
C2, etc) in a #latex document.
This is a requirement of the VU thesis style, but it tool me a while
to figure out how to do it without requiring manual intervention for
each appendix chapter. The simple way is to restart the page numbers
and redefine \thepage to include the chapter letter:
But then you have to put that at the start of every appendix chapter.
So the solution is to add an at-begin-chapter so the above is repeated
at the start of each \chapter. I couldn't get AtBeginChapter to work,
but this equivalent using etoolbox does, thanks to lockstep on
Interesting to see lots of articles today talking about how the
shutdown of Google Reader means "RSS is dead". RSS was alive long
before Reader - it is great software, but it is just one of thousands
of RSS aggregators (see the bottom of this post for more). Firefox
even has a built-in RSS reader with sync between devices.
For those who don't know, RSS is a system sites can use to publish
when they have new articles - you use "RSS aggregator" software to
tell you there is new content and quickly filter out what you aren't
interested in based on the title, summary or keywords. It keeps track
of what you've read and what you haven't, and articles you want to
read or refer to later.
If you read a lot of content each day its an essential tool. With RSS
you don't have to manually go to each site to see if they have
something new. You don't have to wade through all the content
yourself, and keep track of what is read and unread yourself. I
estimate it doubles or triples the amount of sites and material I get
through, mainly because I can quickly filter out what I don't want.
Many years ago I stopped reading sites unless they have an RSS feed I
can connect to (and filter for what I'm interested in) - I don't have
the time to poll-and-scroll sites that don't. I've stopped reading
sites which have poorly implemented feeds too (attention Reddit and
Hacker News, just putting a title and a link isn't useful). Reader
going away sucks, but RSS isn't going anywhere.
I mentioned Firefox's built in aggregator before, but there are many
RSS reading alternatives which operate in utterly different ways. One
is rss2email - it converts RSS feed articles into emails, so you can
read the feeds using whatever you are currently using for email
(filtering it to a different inbox is highly recommended).
rss2html converts a set of feeds into a text or HTML page - you can
run it on your own machine and generate your own page of feeds, or as
a cronjob on your web server and access it from anywhere. Of course
there are also many "cloud" based ones like Reader, many standalone
applications for all platforms, many which straddle the line by
providing local storage and cloud sync. Finding one that fits the way
you work is one of the advantages of using RSS.
A new section of the site showcasing some of my more interesting
work at RMIT (including my summer research project).
Some extra tips added; I still haven't put in the links yet.
New quick script to serve the contents of the sdcard over HTTP.
- Moved the deprecated image gallery to the misc section.
- Removed some of the nested cruft from the sidebar.
UPDATE: Also split off a new .plan from my old RMITCS one. The old one
will be kept online in an attic somewhere for post-erity.
This article is Apple Mac-centric, but its lessons apply to all other
In one well written post this guy has described exactly what is wrong
with so much of the software made today, especially the trend-fueled
mobile/cloud apps that promise to store everything and organize your
life for you.
(It isn't anything to do with privacy; that is a problem, but only
with the storage, not the interface per se).
My favourite part:
"If you want to store data of differing types within a lightweight
organization system, I encourage you to check out the filesystem.".
"The point is that if **you don’t do** some organization of your
data up front, you probably won’t like the ways in which **it’s
done for you** later." (emphasis mine)
New security notes page oriented towards experienced
users who are not necessarily security experts. If you can
use "gcc" and "man" but not "snort" this is for you.
There are some not-entered links on there which will be
fixed ASAP. I plan to add more to this page soon,
Cleaned up the comics I no longer read and added a
section for news sites.
- New codepile script: logfx. Logfx will monitor files for new lines
matching regular expressions and play sounds if it finds a match.
Files/regexes/sound effects are all configurable in flat files.
- More unix commandline tips.
- Improvements to Rastodo:
- Terminal and Android versions are now merged fully again
- bug with --appointments notfiltering properly fixed
- New --two-lines option to split entries so descriptions have
their own line.
- Probably some other minor fixes I've forgotten about...
Recently several smartphone app developers have been widely published
complaining that Android is a "piracy platform" and they make less
profits on Android. The former is blatant doublespeak (conflating
freedom with Piracy) and while it is certainly much harder to make
money on Android it is not due to piracy - it is because of the huge
collection of free apps, which flourish due to the freedoms that
Android developers enjoy.
By "Piracy Platform" what they really mean is that you can install any
app from any source on an Android device. This is not a deficiency of
the platform, it is a feature - Android users have control of their
own device and are not limited to the Official Store. This feature
gives users the freedom to do what they want with their device.
Developers can still confirm a user has paid for an app by using the
Application Licencing Service
This can also limit an app to running for a particular time limit or
on a specific device. While this is enabled by default in the 4.1
version of Android, it still works on devices running the ancient 1.5
version - if the developer bothers to use it. From what I have been
able to discover, the most vocal piracy complainants are not using the
licencing service for their apps.
With an Iphone/Ipad users are limited to only the apps which Apple
allows them to use, and they can delete apps - even ones you have paid
for - from your device at any time. (Would you limit grocery shopping
to a single supermarket, or let a single publisher decide what books
you can read? When they stop publishing a book, would you let them
enter their house and take it off your bookshelf? But we should return
to the issue at hand: piracy.)
The freedom to install unofficial apps on Android does make it
slightly easier to pirate apps. The user doesn't have to jailbreak
their device, although someone still has to crack and remove the DRM
on the Application Licencing (if the developers used it). This still
doesn't prevent good apps from being profitable.
The obvious analogy here is Microsoft Windows or DOS: it is even
easier to install software from any source, copy protection is
somewhat effective but often cracked by Warez groups and piracy is
rampant, yet many developers have made millions writing games and
other software for Windows.
The real challenge on Android is not piracy, but free programs that
fulfil the same function. If you make a good game, worth shelling out
$5 for, some people will pirate it but many will pay for it. But few
people will pay $5 for a unzip program, music player, simple puzzle
game or a moderately-useful-but-trivial-to write widget when there are
completely free programs which they can use instead.
This is why it is hard to make money on Android or Linux development -
to make it worth buying your software has to do something very useful
to the user. Something that the user can't get from free alternatives.
This requires significant work because there are many high quality
free alternatives on these platforms.
Battery Widget Pro will not sell for $1 because there are dozens of
free battery widgets. Neither will a flashlight app or a simple
solitaire game. Something rare and useful, like a good Word Processor
app or music composition software, and droid users will happily pay
for it because there are no free alternatives. But if it only takes a
little effort to write, another developer will make a free version.
Why are there so many free apps on Android? Because developers have
more freedoms and lower costs, so there are more Android developers
willing to make free apps. It costs nothing to get the Android SDK,
and you can run it on nearly any computer and operating system made
this century. You can use common general purpose programming languages
It costs nothing to publish an Android app, you can do it on your own
website. Many developers for Android and Linux write software for
their own use, and because it costs them nothing to publish, they make
it available for anyone to use for free.
On the other hand, there are no truly free apps on Iphone/Ipad. The
developer is paying at minimum $99 each year for their developer
licence, and (unless they already have one) $1200 or more for an Apple
brand computer which is the only type which can be used to develop for
Once the developer has spent over a thousand dollars to get set up and
ongoing costs every year they will want a return on that investment.
Thus there are fewer free apps. The ones which cost nothing to install
usually make money for the developer in some other way, via
advertisements, in-app purchases, encouraging people to buy from a
related store or service (e.g. Coles and Woolworths shopping cart
apps) or by siphoning contact data and other useful information for
marketing and advertising purposes. Security researchers have found
nearly half of Iphone apps send personal information such as location
and address books to the developers or their marketing partners.
Of course, the lower barrier to entry on Android also means there are
many low quality apps around. But due to a simple user-driven rating
and review system the chaff is easily filtered away.
Finally, just take a look at who is making money on Android, and what
they are providing to their users. Documents To Go and Quickoffice are
in demand and very profitable. Angry Birds, of course. The original
Final Fantasy III can be played in a free emulator but over 50,000
users spent $17 on the official app version. The brothers who
developed the Star Trader and Cyber Knights RPGS provide a free
version of their games but make a profit on the "Elite" version.
-- Dylan Leigh
- Added my Todo Graph to the main site.
- A new Unix Notes page, for assorted command and shell tips (e.g.
random find/awk/tar tricks).
- Minor corrections to various other notes.
- Warnings about the age of the Image Gallery. :)
Tryng to improve my workflow by labelling my mail in Mutt.
(Gmail and some other systems might call this "tagging").
I prefer to keep most of my directly addressed incoming mail in the
one inbox, so I can see it all at once. However, I need a way to mark
some messages as for DP, teaching, research, social and other roles.
Mutt has built in support for searching, sorting and filtering on the
X-Label header (using ~y<label> as the pattern) and showing it on the
index view (add %y to the index_format) but there is no handy way to
edit it from the index (without editing the whole mail in the text
This is the sort of thing that should be easy with a macro that pipes
it through formail, but isn't, because I can't get mutt to replace the
old message with the new one. When I've slept on this hopefully the
solution will come to me.
Speaking of mail, as mentioned in my last post, My student email
addresses at RMIT (s30...@student... and dleigh@cs...) will expire in
a few days. Update thy ~/.mail_aliases.
Attention all Humans:
My student email addresses at RMIT (s30...@student... and
dleigh@cs...) will expire soon. I know a lot of people are still using
My RMIT staff account is checked daily and will be active at least
until the end of the year; my home accounts should be active
indefinitely (although they are only checked once or twice per week).
Proof that Evolution only produces improvements very slowly, over many years.
(Also, it is not a product of Intelligent Design)
So long waiting in line at Safeway I had time to rewrite my ModeSelect
app (from my phone). This has been added to my Android code pile on my
Also just added a short script (droid-netifs) to show the IP and name
of all active network interfaces.
------==========\ Dylan Leigh - www.dylanleigh.net /==========------
Generated Tue, 21 Feb 2017 01:44:49 -0800
Copyright © 2002-2014 Dylan Leigh.
[HTML 4.01 Transitional]