Posted on Sun Jul 13 20:30:00 CDT 2014 by Savanni D'Gerinel
My old set of flags finally bit the dust. There is damage on one of them that I can't stop, and it gets worse every time I fly them. This set took only an hour to cut, fold, and color, and then they had seven hours for the color to cure. They dried in less than fifteen minutes. I hope to get my hands on a sewing machine some time in the next couple of days and want to have these ready to fly by Friday.
In case you are wondering why there has been no new HDR photography in a long while...
Luminance-HDR is now totally broken. Every image I generate with it has a precise square of discoloration in the middle third of the image. This is not a problem with my camera, as I have actually gotten the same results from photographs I took a few years ago on my old camera. This is the most mysterious of all bugs, and it has ended my patience with Luminance.
My own software is not complete. I have the math problem set up, but get back the wrong answer every time. I may have to start munging the answer to get it right. In the interim, I'm going to track down the source for pfshdrcalibrate and get it compiled. I'll revive my old toolchain, which really did serve me well for a couple of years.
Posted on Wed Jul 2 10:45:00 CDT 2014 by Savanni D'Gerinel
Plans change. Mine are changing generally faster than I like. But for the last year I have been having trouble figuring out where to go next and what to do next. Admittedly, the simply physical "where" I have a handle on in that I am heading for Oregon, but whe bigger picture "where" was invisible to me.
Today, a change of plans.
I will, instead of heading to Oregon, spend this autumn in the Hacker School in New York.
You can't see me, but I'm trying not to just explode right now.
I have a whole checklist of things that I have to do to see to it that my life is properly preserved while I'm gone. I'm not taking everything with me, and it occurrs to me that this may be yet another opportunity to figure out what is important for me to keep and what else I should get rid of.
I also think I do not want to continue living out in the woods of Texas for the remaining months. I'm paid up through the end of this month, but I kind of think that I want to spend August and September actually inside of Austin. Something I will need to look in to, really.
Posted on Sun Jun 29 13:45:00 CDT 2014 by Savanni D'Gerinel
My email address is not secret. I have it on my About page. But there is a problem. Most everyone I know uses Google for email. A few people use Yahoo. I can't reliably send email to either group. There may be other groups, too.
Almost every email that I send out is a multipart email. The first part is the actual text of my email. The second part is a cryptographic signature that can prove almost absolutely that I sent the email.
If you are a Yahoo user, you likely see only a blank email with an attachment. Yahoo has never handled the PGP/MIME standard correctly, even though RFC2015 is almost twenty years old and RFC3156 is almost fifteen years old, I've known Yahoo users who assumed that I was sending out viruses.
If you are a Google user, you likely will not see my email at all. Google, whether malice or incompetence, has decided that PGP-signed email is spam, and so almost every email I send to a Google user ends up in their spamboxes. Sometimes, through text messaging or face-to-face conversation I can get such a user to dig through their spambox and then try (sometimes unsuccessfully) to train Google that I am not a spammer. But that generally makes it impossible for me to initiate contact with a Google user.
I have no idea what Hotmail and other web services do. I can only hope that maybe they are smarter.
The thing is, I am following internet protocols. I am following long-standing well-established standards. And these services punish me (and you, for that matter) for using these protocols that make the internet work better for everyone.
So, if you wish to communicate with me via email, please take pre-emptive action and do whatever it is that it takes to make your service believe that emails from my email address are valid. Given that there will soon be far fewer ways to contact me on the internet, this is rather important.
Posted on Fri Jun 20 18:00:00 CDT 2014 by Savanni D'Gerinel
Described to me, frequently, as "Gay Pride as it was before the corporates got it". I've never been to Pride, only to this. And this is fun and awesome. Maybe I can instigate something like this in Oregon if Pride turns out to be as bland as everyone says it is.
Incredibly difficult shooting conditions. Moving crowd. Dim light. Moving me. It's a miracle I got anything that I'm willing to publish.
Posted on Sun Jun 15 16:00:00 CDT 2014
I spent this last weekend at Texas Linux Festival here in Austin. During the event I went from session to session, both learning new things and supporting my colleagues at Goldenfrog in discussions of internet privacy and overreaching surveillance.
I also contended with the distressingly short battery life in my laptop. It likely has little to do with the battery and more to do with the way I specced out the fastest multi-core processor I could afford at the time. I'm in the market for an ultraportable with nine hours of battery and highly linuxafiable. I may even be able to re-acquire my old Acer Aspire One and get a five or six hour battery into it.
So, I rocked it old school with a fountain pen and paper notebook, neither of which is vulnerable to attack from the
NSA agentsscript kiddies that must abound at all tech conferences.
However, I found most of the convention dominated by business interests. These interests are not bad at all, and I certainly appreciate the amount of business support that has gone into Free and Open Source software in recent years, but I also find it bland.
Cloud this. Cloud that. Put everything in the cloud. Use the cloud to manage your data center, handle your immense data needs, ensure your customers against data loss, and so on and so forth. But, how is this a convention for me? I am a lone developer living out in the woods on a tenuous internet connection. I am not CERN or Google. I do not have petabytes of data laying around on thousands of systems, all needing to be indexed and analyzed. I do not have terabytes coming in every day. I just have one (or sometimes two) computers on which I want to work without having to depend on or ask permission of the Cloud.
Obviously, I'm a bit quirky.
You know what I want to do? I want to download, index, and search the Haskell documentation independently of the web. I want to cache web pages, such as really helpful Stack Overflow answers and Postfix tutorials. I want to walk into a room containing my friends and hand them a file without having to read them some arcane URL in a third party service. And I'd love to be able to take a small, low-power, portable computer into Austin, download all of my podcasts, and then have them seemlessly work on the larger computer plugged into the sound system when I get home.
I do not believe that I am the only one in the US with these particular quirks. Technically adept, technically avid, insistent on holding my own data locally so that I don't have to depend on an internet connection to work. And, finding just how hard that is today.
FOSS is about empowerment, and part of empowerment is about choice. Both the ability to choose and the will to choose. I'm a programmer. In the words of the great philosopher Jayne Cobb...
I'll be at my keyboard.
(or something like that. :))
If you ever want to know what I'm working on, check out my Bitbucket account. I don't always have time to write about what I'm doing. And I may sometimes spend weeks working on proprietary code, only to come back to spend weeks working on the open source code that supports it. But the code is where the action is.
Posted on Mon Jun 9 08:00:00 CDT 2014 by Savanni D'Gerinel
Places have memories. Ages. No matter where you go, beings have walked there from the time the place has formed. To call a land "virgin" is to overexault ourselves.
Yet, one thing we do well is to erase that history. To eliminate the natural growth and decay. To scrub away the color and texture and try to pretend that we are the first to arrive, and that all that came before never happened. And then, one day, we leave as well, but leave our own mark on the land.
Posted on Fri May 30 16:15:00 CDT 2014 by Savanni D'Gerinel
Assume that you have two datastores, both of which simply contain a collection of files. This is actually how my website works. I have a datastore on the server and a datastore on my personal computer. Whenever I want to make an update to data on the server, I make the update on my computer, review it (I'm running my website software locally), and then run a command to push it to the server. Not until I push the files to the server do I actually need an internet connection.
This is actually really straightforward, and it works very well for document-based collections like a weblog or photographs. It really covers a lot of the data that any individual uses on any given day.
With a scheme like this, databases never get replicated. The files contain the Data Of Record, meaning that any application accessing the files needs to update any databases it keeps in accordance with changes to the files. If a file disappears, the application needs to notice and remove it from the database. If a new file appears, again the database needs to update the database. If the files get copied to a new system and the database gets left behind, there should never be any data lost.
This method is actually very straightforward, though I do not yet know how far it can scale. It may actually not scale beyond a few hundred files. I recently purged my photo database, and then instructed Digikam to reconstruct the database. Until that database was reconstructed, no tagging was available on the RAW files that I keep around, so I was unable to look at my "pool of waiting photographs". The reconstruction process took about four hours. My collection has 23,000 photographs (maybe I should do some cleanup of the old ones that I'm never going to process...).
Complete purge/refresh of a database is extremely expensive. Incremental updates tend to be very cheap. So, for now, I'm going to advocate this method for document-based data.
Posted on Mon May 26 08:00:00 CDT 2014 by Savanni D'Gerinel
I want to build an omni-tool. Or, really, a non-sci-fi equivalent thereto. We already have all of the parts for something that effectively is one. Ignore all of the holographic user interface that appears in all of the Mass Effect games. Also ignore rather silly things like nano-fabrication and whatnot. Those aren't the core of an omni-tool.
At it's core, an omni-tool is a portable, nearly wearable computer, that remains useful even when disconnected from all networking functions. In the modern day, only cell phones are always connected, but they don't have unlimited bandwidth and you typically cannot share your cell connection with a device such as a tablet.
So, when I talk about an "omni tool", I am actually talking about an information appliance that works well in a disconnected mode. For that, we need software that...
- Helps me to select what data to cache locally from my home datastore (i.e., what music, books, documents, etc, to copy to the local machine)
- Helps me to select what data to cache locally from The Internet (i.e., what static web pages should always be cached)
- Allows me to enter new information into the local cache, and have that local cache integrate the data into my home datastore when I get access again. Perhaps it requires some interaction on my part, but it should be pretty straightforward. My concerns should be about deciding how to resolve conflicts, not finding the data that I want to integrate
My Use Case
Always be use-case driven. That really helps out with determining an architecture.
Given my current life, here are my use cases for when I carry around just a tablet instead of a full computer:
- download podcasts and videos to my tablet
- carry a copy of the documents that I refer to
- have my todo list and calendar
With podcasts and videos, "integrating" them involves making gPodder on my desktop machine recognize that a new file has appeared, even though it did nothing. Videos are easier since I have no management software, just a particular place to put them in the filesystem. Todo lists and calendars are harder, though I have todo.txt (which only synchronizes through Dropbox) and Google calendar (which only synchronizes through Google). Technically, both of those options suck because I have to literally go online in order to update my devices.
I've thought about these problems since about the day that handheld devices started becoming common. I have, on occasion, come close to solving the whole thing. But, since it has never actually been a use case for any application I was getting paid for, I have never actually implemented useful code to solve this.
Be that as it may, for the rest of this week I am posting additional articles to further explain what I am thinking of. Maybe some solution can become clearer.
Posted on Wed May 14 00:00:00 CDT 2014 by Savanni D'Gerinel
At one time, I made several efforts to put my work up in galleries. I have since stopped, mostly due to distraction by other tasks, but I still have prints of some of my work. Here are four of them, and one of them is going in rather short order into a raffle to help fund the Toltec Center of Creative Intent. I am posting them here so that others can take a look and give feedback as to which one I should donate.
Three of the pictures are printed in 8x12 format, one printed in 8x10 format, and all framed and matted to 14x17.
Posted on Thu May 8 19:00:00 CDT 2014
Last night I had the opportunity to talk to a local community here about gender binary.
I wrote this big huge lecture (probably fifteen minutes of talking), but then I didn't actually use it. Most of the attendees were people who I had invited as support network. I did not really want to preach to the choir. Maybe I should have so that they could help me refine the talk. Instead, we watched a video, I talked about creating safe space, and then we went into discussion.
In any case, I'm including the full text that I was planning to use. I will use it again for future such opportunities and may simply use it no matter how large the group is. I even noticed that I never actually finished the text, so I will fix that, too.
I will update that page as I refine the text or find new resources. Here, however, is my current list of resources for research.
- The Genderbread Person, v2.0
- A List of LGBTQ Terms
- Suicide Attempts amongst trans and gender non-conforming people
- Understanding the Complexities of Gender, by Sam Killerman
- Ending Gender, by Scott Turner Schofield
- Genderqueer Identities
- A list of Genderqueer Identity Terms
- About Trans*
I'm thinking that I myself will need to start becoming a resource on gender non-conformity. I will see how that goes.