Has WiFi Peaked?

So in the course of The Great Media Experiment, we've noticed a significant degradation of the Apple TV's ability to stream data from the computer upstairs.  Forget HD stuff, we can't reliably stream DVD rips.  One has to start the video, pause it, go away for ten minutes to let it cache, then start it.  And even then, some evenings that isn't enough.

The iPads are similarly affected, but that's usually written off to them being tablets with obviously inferior wireless equipment.  Why obviously?  Well, because it is bad, of course!

The real, real culprit is probably this:

Well there's your problem

Yeah, eight networks overlapping mine.  Nice.

(Actually, looking at that makes me wonder if I can solve a bunch of my problems by shifting to channel 11, since there are fewer coincident networks there...)

So yesterday I spent a bunch of time drilling holes in walls and floors and I ran a wire from the basement into the Apple TV.  And wouldn't you know it, all of our problems with streaming are gone.  Looking at the activity graphs on the computer upstairs, it looks like it can now stream data at 40Mb/s whereas over the WiFi it would plod along at 1Mb/s.  Huge difference.

I also discovered that my laptop can do a speedtest.net and show a 15/3 connection over the wireless.  Connected to the wire? 50/3.

It has a bit of the boiling frog problem.  When I started with this WiFi in the home thing eight years ago or so, I had a router in my basement.  The theory was that it would project the WiFi signal more or less straight up, and the concrete foundation would "hide" the signal from the rest of the neighbourhood, and to a certain extent hide the rest of the neighbourhood from my router.  It was all kind of academic, since I remember that I could see exactly one other wireless network from my laptop, and that weakly at best.

Now here we are, I have changed WiFi routers three or four times -- each time when the signal reception was getting unusable, I just assumed the router was dead or worn out and got another one.  The router has also been moved around the house -- from the basement, to the second floor office, to the shelf in the front room (a location I picked because it was the only network jack on the main floor).  The router's current location is less than 10m from the Apple TV with no significant obstruction between the two devices (one empty desk and the side of the entertainment unit -- no walls) and the throughput is still crap.  The iPads can barely surf the web from the couch in the living room.

I've even contemplated a wireless mesh network for the house -- multiple radios on each floor.  And frankly that seems just overkill ridiculous to me.  I'll be blasting out so much signal that my neighbors won't have any choice but to respond in kind.  And we'll be right back to where we started, except with more money spent.

5Ghz doesn't look like a solution because it doesn't work even half as well as 2.4GHz -- and given the choice, most devices will chose the 2.4Ghz connection anyways.  And looking at the InSSIDer screen above, it appears my neighbors have come to the same conclusion, as there's practically no 5Ghz networks in use.  Possibly the prevalence of 5Ghz cordless phones is what is killing 5Ghz WiFi.  I know we have 5Ghz phones.

I'd have to do the mesh solution to make 5Ghz work because it demonstrably doesn't work between floors.

All this makes me wonder if for the average person, WiFi might have peaked.  So many people have so many access points and devices that the spectrum is so congested.  People tolerate it because their tablets or phones are mobile devices, and therefore inferior (see my logic above) or because their laptops are running Windows and therefore obviously slow.

Or because their internet is slow.  (But that's a separate blog post.)

Overall the experience is so much worse than it could be.  Eventually people will stop putting up with it.  I wonder if my return to the wire is the beginning of the end of the WiFi wave in the same way that my router was the beginning of the same wave.  

As for my network, I'll probably move the router next to the Apple TV -- not for wireless streaming, since the wire is superior in every way -- but so that the iPads work better in the area where they primarily get used.  I'll try the channel change I speculated on above.  And I'll look into another router for the office upstairs.  We'll just have to see how it goes.

Review: Burlesque

Plotwise this is a terrible movie. The main protagonist's trajectory is only up, up, up. The only crisis in her trajectory is an incompetent professional rival who merely provides a platform for our protagonist to show how wonderful she is; and will she find a boyfriend. The main crisis in the film, the club's survival, is driven by a supporting character.

If you ignore that, it was loud (which might have something to do with the TV it was displayed on) and choppy (which might have something to do with the network it was streamed over).

Review: The Hobbit: The Desolation Of Smaug (IMAX-3D)

IMAX is really big. Too big, if you sit too close. So if you have to sit further back to make the big screen smaller, what's the point?

I preferred the HFR-3D that we watched the first part in, even if it seemed a little "darker" (visually, not story-wise). When you have a gigantic screen at the standard frame rates, fast movement is visually blocky, as in the monster is here - here - here, not flowing from point A to point B. HFR gives you more frames so there are more places for the monster to be in the same period of time, so the movement has to be much faster before it breaks down like that.

I am still not convinced that 3D is anything more than a gimmick. This one made me dizzy with a couple of camera moves though.

And oh yes there was a story.

My biggest complaint: so far there are no stakes for the heroes. Nobody has died yet. I mean in LoTR 1 you had Boromir step up and take one for the team and you got the message, hey this isn't some fairy tale where we're all going to live happily ever after, some of us will not make it out. In The Hobbit there are thirteen dwarves at the start of the journey and there are thirteen dwarves at the end of movie #2 and the most peril they've been in is one of them got sick. I know this is supposed to be a children's story, so all of our plucky heroes need to make it, but since we're showing gigantic spiders I don't think this is a children's movie even if it is a children's story. Orcs are being mowed down like storm troopers; let's loose a dwarf or two. In the noble sacrifice for the rest of the group, yes, sure, but come on.

I got a little tired of the action sequences, some of them were more than a little contrived; nearly all of them went on too long. Say, for example, the bit where one of the barrels bounces out of the water and across four or five orcs.

Speaking of contrived, I think that word sums up Legolas' inclusion in the story. Word is that Vigo Mortenson declined to reprise his role of Aragorn for The Hobbit on the grounds that it wasn't canon. Well, maybe Orlando Bloom needs the money, and frankly who is he to say no if someone wants to pay him?

Jenn wanted to know that since Gandalf knew for certain that Sauron was "back" so to speak, what was he messing around with for sixty years before LoTR's start, and why was he so horrified that Sauron was back in LoTR?

And-oh-by-the-way, the One Ring is a universal translator for evil?

Overall, this movie is weaker than the first one, even considering that part of the strength of the first one was that feeling that ok we've sat through all this the next two will make it all worth while, and frankly so far it isn't. This movie was too long, and together the first two movies are way too long. I'm getting the feeling that once the third movie comes out, someone could make a 120 minute supercut of all three together to tell a tight and coherent story without making us sit around for eight hours.

Prostitution

I'm not terribly convinced by the legal argument that prostitution shouldn't be illegal because having it so makes it dangerous for these involved in said illegal activity, but that's probably why I'm not a Supreme Court judge.

Git Is So Terrible It Is Awesome

Today's SysAdvent is hilarious:

14 Tips For Git Giddiness In 2014

It's hilarious because it basically boils down to git is so awesome that it can commit every sin that would get any other piece of software summarily canned and you'll keep coming back for more.

It features such pearls of wisdom as:

• You'll be glad you stopped relying on default command arguments, especially when they change in a major release.
• Previous version control systems would add conflict markers in the file and make you resolve them, but without operating against the repository. (and here's how to make Git work the same way.)
• Do Not Store Artifacts in Git (Even Lil' Ones)
• Submodules are Horrible... Except When They're Not
• One of Git's most useful features is the ability to add commands to its lexicon, via the alias feature. Learn to love this feature: it allows you to create shortcut commands for complex operations.

That one is my favorite -- after telling you to avoid default behavior because it will break in future revs, the author tells you to create your own defaults, which will break in future revs because the pieces they depend on will break. Plus it will mean that not only do you need a particular version of git to work the way you want, you'll need a whole environment.

• Be Careful What You Read On the Internet

That one is good advice. One could even apply it to both this and the linked article.

• The attitude that everyone must obsess over the intricacies of a single counter-intuitive tool required to get work done does nothing to help Git's adoption.

One might argue that Git being a counter-intuitive tool is a bigger hurdle preventing Git's adoption.

• Your Team Needs a Git "Language Lawyer"

Ha.

• Make Time Early On to Discuss and Decide on Workflows

I describe this as "the wiki problem". Because there's no structure imposed on people, people can do what they think they want. Which is liberating. But the problem is that if you have more than one person using the tool -- and sometimes not even that many -- the conflicting ideas of workflow (or structure, in wiki terms) turns the whole thing into an inconsistent mess, one that only increases exponentially in complexity and depth as more people are added.

In summary: Git Is So Awesome It'll Make You Want To Kill Somebody.

The bottom line to all this is that version control is hard, even though we've been doing it for ever -- sccs dates back to 1972, which is before most of the kids today can remember.  Of course, these kids "discovered" version control with git, so they think they invented it and that it is all new.  Us old fossils know better.

And the best part -- we can watch the same thing happen to these kids.

Even Better

FIVE zeros.  Just sayin'.

I think that's good

OK, I guess we can make it into orbit now.

Let's Play: systray

Who can identify all of the icons in my Win 8.1 systray?

Everyone can play!

Why is it that every program you install seems to need an icon displayed to tell you that it is actually doing something?

Answers when I get around to it.

Review: Thor - The Dark World

Pretty much what you'd expect for a comic book movie. Villains which are less than credible, lots of "magic", and plot holes big enough to park other movies in. But smashy smashy fun. I'm not the Loki fan that my wife is, but he's ok. And frankly at this point three movies in you have to respect the writer/director/producer's restraint from doing a M.C.Hammer "hammer time" joke.

This Is Not The Droid I'm Looking For

Me: I'd like a new iPhone, please.

Salesdroid: iPhone is over,  you should get the Samsung Galaxie 4.  Cheaper and a better phone.

Me:

A) Will it run my iPhone/iPad apps? – no
B) Will it play my iTunes-purchased music and videos? – no
C) Will it get firmware and security updates in a timely manner? -- impossible to predict with 100% certainty, but history strongly suggests: no

Keep your android kool-aide

Four Zeros Of Orbital Eccentricity

So due to copious spare time, the KOS project is grinding along slowly.

It turns out that the hardest part of the launch sequence is getting a halfway-useful circular orbit.  The math is hard to figure out, and it is difficult to work out a spit-ball method of trying to do it reactively.

So, I cheated.  I abandoned 0.8.4, which had a bunch of exciting bugs, and am now running 0.9.2 instead, which has different exciting bugs, but fewer ones appearing in the currently useful command set.

More to the point, I also found these scripts on the KOS wiki -- a useful, if occasionally misleading, resource -- and with a bunch of tinkering they get the job done.

My changes are not that significant.  I threw away the warp handling because it didn't work for me.  The revised node execution script now tells you that you are in a warp-permitted window, then unwarps you forcibly 60 seconds before the node should be executed.

The net result is this:

Four zeros of orbital eccentricity.

KX-06 is in a 110x109 km orbit.  That's pretty damn good.

The KX-series is the same launch and orbital vehicle as the KT-series.  I called them KX instead since I am experimenting with someone else's code, not testing my own.  But that's ego for you.

As you can see, since the last update, I've been firing off rockets like crazy, with varying results:

...but to be fair I should mention that some of those were experiments with geo-synchronous (kerbo-synchronous?) orbits that didn't pan out for whatever reason.  KSP reports 19 probes in orbit at the moment.  And there have been a ton of launches that didn't pan out.  Over all, what with reversions and all, this has probably been in excess of 60 launches.

But it's all for science, right?

I've also been strict about by debris.  As of right now there are only three sets of orbital debris -- one for KX-06, which is still in its launch ballistic track,  and two tied to the kerbo-synchronous attempts which will take several days to return back to Kerbin.  End result is that I've got a lot of stuff in orbit, but nothing that is junk.  Well OK, nothing that is all junk, anyways.

I feel bad that I cheated, that I wasn't smart enough to figure this out.  Part of my problem is that I didn't know what KOS exports in terms of usable values, and the ones I was playing with for the KT-series just didn't have the information in them that I needed.  In reading through the code I've learned enough that I could probably code a reactive circularization module -- but frankly doing it "correctly" with calculated values is a much better way to do it.  And I've stared at enough wikipedia pages on eccentricity and semi-major-axis et al that I think I understand the math in the code.  I might re-code the module myself to ensure that I really do understand it, but I'm not sure that's necessary.

So what's next?  The original plan was to use the 100x100 orbit as a "staging" orbit and then calculate ways to transfer probes to more useful altitudes.  I was going to write a series of programs to automatically put satellites into kerbo-synchronous orbit with particular destination slots, just because.   But that was before circularization started to kick my ass.  So for now I don't really know what I'll do.

But frankly there's so little reliable spare time these days that I probably have quite a while before it becomes an issue.

KT-10 makes best orbit yet

KT-10 makes a self-guided launch into a 142x99km orbit.

Standard "Baniff" lifter

Program complete

This project is using the kOS plug-in to manage the launches.  Everything is done programically.  The only inputs I make manually are view changes and the action group to command the solar panels and antenna to deploy.

Surprisingly, or at least surprising to me, is that the hardest part of the launch is the circularization burn.  If you just whack it at apoapsis, you end up with a hugely elongated orbit.  Early KT-series probes even achieved escape velocity from Kerbin.  There's got to be some math I can do to take my altitude, velocity, and vector, and calculate both my current orbit and my desired orbit -- and therefore the circularization burn to transition between the two.  Right now I'm trying to balance my altitude, apoapsis, and periapsis dynamically and the 142x99 is the best I've achieved.  As you can see below, the KT-series has come a long way:

KT-1 through KT-9.
KT-9 highlighted.

Review: Escape Plan

Not nearly as terrible as it should have been. The predictable plot problems, but really who goes to see these movies for plot? Also the wrap-up at the end was a little weak. But it hits its target, as long as that target is set low enough.

Review: Rush

Excellent. Not sure that it will surpass Grand Prix as the seminal Formula One movie, but a very well done effort.

People Can't Drive, A Continuing Series

I find it fascinating that this:

...appears the same day that this article does:

Executing the "zipper" merge. Road work often reduces two lanes of traffic down to one. In these situations, American drivers typically merge into the right lane as soon as possible and form one long line. The main reason they do this is because people think it's bad behavior to stay in the left lane and merge late.

In fact, says Vanderbilt, traffic would be much better off if cars stayed in both lanes then merged at the very end, one by one, like a zipper. It's safer (fewer lane changes), it reduces back-ups (often up to 40 percent), and it quenches road rage (still on the rise).

The zipper merge is used in Germany but can't overcome its bad reputation in the United States. A trial in Minnesota failed because drivers wouldn't stay to the left. They were too nice.

Personally, I usually stay in the vanishing lane and drive right up to where it ends, then sit there and wait for someone to let me in.  Someone always does.  Besides, what other purpose would having that lane open to that point serve?  If one was only supposed to drive in a lane that wasn't going to end/exit, the entire highway would be one lane.

Learn to drive, people.

Review: Riddick

Cheap night, best of a bunch of bad offerings, but I think I still overpaid. The machinations to get him out of the end of the previous one were a bit contrived, the entire pre-station piece is more of an introduction or prelude. After that it settles down to Riddick-being-Riddick set in a world of implausibly horrific monsters. The previous one was better, although one starts to suspect that it had a lot of other moving parts around the core Riddick-being-Riddick and the end result was an accident.

In which we go to Moho.

Yeah, G+.  But what are you going to do?

Review: Star Trek - Into Darkness

This movie was easily the worst move I have seen this week. And that's considering that I watched The A-Team this week.

Review: The A-Team

A terrible movie, redeemed somewhat by the fact that nobody expected it to be anything but a terrible movie.

Review: Pirates Of The Caribbean -- On Stranger Tides

After typing in that long a title, I think it needs a longer review than the "That. Was. Terrible." reaction that I had when it was finished, and this it that longer review.

Review: Oblivion

I liked it but I'm honestly not sure why.

There were problems with the way the drones worked and the mechanics of the broken moon. But I liked it, even if the end was perhaps a little lower key than we might otherwise expect.

(Update: I paid to own it in HD, so I must like it.)

Review: The Wolverine

Pretty good for a comic book movie. Not knowing the cannon made it easier to follow without guessing what will come next. Plot holes can be handwaved away with the "its a comic book" excuse. The credits teaser had problems.

Ionic Mk.3 Arrives At Dres

Mission log:

(MET Day:Hour:Minute; IB=Ion Burn #)

285:06:24 IB17  +2h11m Match orbital speed with Dres 1693m/s

• Gigantor performance at 35.5Gm: 14.79 -- so a single one is sufficient to drive an ion engine at least out to Dres.
• We are 500Mm away from Dres.  I think this means we pooched the Hohman intercept.  On the other hand that's like a 1.3% miss.

351:12:50 IB18: +3m15s 45m/s adjust intercept orbit

• Closest approach will now be 7Mm in 1y 12d.

594:18:04 IB19: 7m 35s, 99.9m/s Set up for Dres encounter

• Encounter Periapsis currently at 700Km.

The tiny dot just above the probe is Dres

992:14:28 Dres encounter. Encounter periapsis set for 580km.

Orbital Insertion

994:10:10 IB20: 5m52s, 78m/s -- insertion into 600x10000km orbit around Dres

994:10:17 Engine shutdown, welcome to Dres.

Engine shutdown, welcome to Dres

So aside from there being a lot of waiting for ion burns and for planets to get into the right position, that was an interesting exercise.

Now I have to decide what to do next. Probes are interesting, but I have a desire to do a lander again.

On the other hand, we still have 4800m/s of delta-v available in the Mk.3.  Maybe we should try to push it out to fly past Jool.

Isn't it Ionic?

...just a bit.

So with the wrapping taken off of 0.21, I decided to play around with ion engines this time around.  Why?  Well, I still have a mental block about killing Kerbals, so I'm reluctant to play with "manned" missions.  This does rule out a lot of the potentially fun stuff.  So being the contrary so-and-so you all know, I decided that I'd pick something that I don't read about much in the G+ group.  I'd like to do something new, rather than trying to repeat what others have already done.

And so: probes with ion engines.

The original title of this article was Ionic Mk.3, from which you can infer there was a Mk 1:

Mk 1: Fancy girders.

...which got sent on a deep space mission to Eeloo for the usual no good reason that I'd watched a video about someone going there.  Somehow the upper stage of the rocket I assembled had the Delta-V required to inject into a transfer orbit and to circularize out there.  Once an intercept was set up, the upper stage was discarded, and I discovered that the solar panels I had put on were insufficient for the task of powering the ion engine way out there.  After the Eeloo flyby, I set up an intercept with Jool, but I only get about seven seconds of ion run time out of the batteries before I have to charge, and a quick back-of-the-envelope calculation showed I'd need about 50 of these to get the required five minute burn needed effect the Joolian intercept.  Now JPL can do this, but I estimated this would take three or four of my real hours to do this since there doesn't appear to be a programmable automation system -- what I'd need would be "fire engines until charge is down to 10 units, then turn engines off until fully recharged.  Repeat all that until the required delta-V was effected."  This didn't seem that interesting to me, and I abandoned it.

So it is still floating out there, but unused.

And of course there was a Mk.2:

Flashy flashy.

This time I exploited a clipping problem to stack two ion fuel tanks inside the battery pack, and attached the ion motor to the base of that.  So it is a lot more compact than it might otherwise be.

And, being an inquisitive KSPer, I've watched the Kickstart video for the micro-ion satellite kickstart.  And that animation showed how you gain escape velocity with an ion engine: very slowly.   As you pass your periapsis, you burn the ion engine briefly to push the apoapsis out a bit more.  And you repeat on the next orbit.  And the next. And the next. And (you get the idea) the next.

I picked a three minute burn for the usual no good reason.  That gives me a burn at 1m30s on either side of periapsis; I'm sure some detailed math would show what the ideal burn length time was, but I don't know how to do it.

So, for ten or eleven times, I did this:

A bit more this time, please

I flew the first three or four totally manually.  I tried to just point the probe in the direction of travel for the whole burn, eyeballing the MET clock.  Probably not the most efficient way to get around.

After doing this a few times I did figure out how to get MechJeb to help -- I set up the maneuver node manually, adjusting the thrust so that I had three minutes of burn time.  Then I told MechJeb to fly the next node.  So MechJeb would warp to the node, orient the probe, do the burn, then return control to me.  So it was tedious, but doable.

Then eventually the elongated orbit (I'm sure there's a fancy name for this type of operation) crossed the Munar orbit, and something wonderful happened:

Step up and claim your free ejection. 

The Mk.2 return's swing back to Kerbin brought it past the Mun, and suddenly we got a free ejection from the Kerbin sphere of influence.

Well I'm not one to say no to a free ejection... but as you can see the trip past the Mun was actually through the Mun.  Fortunately a lateral ion burn corrected that problem.

Making a close approach a little less close.

...and just to show off, I made the closest approach 10km from Munar datum.  (I didn't have the nerve to make it any closer.)

Sorry, Ghostrider, the pattern is full.

So now I've got it ejected from Kerbin.  I didn't want to go into deep space because Mk.1 showed me that the ion engine might not work well out there -- even though the Mk.2 has way more in the way of solar power collectors.  So I tried to go down instead of up.  Eve is the next target, so I settled on getting into an equivalent orbit.

Problem is I'm totally in the wrong place.  What I need to do is change one of my -psis'es a Gm or so and then wait for an appropriate encounter.  Well that'll take time, so I've pretty much left that one as is as well.

So for today's exercise I decided to add lots of Ion Fuel, just to see if it would be usable.  So this is the Mk.3, which only has one light, and has eight Xenon tanks instead of one or two:

Beauty shot.

Navlight for the usual stupid reason that if I have to look
at this for hours, I want it to be visually interesting.

Eight ion fuel tanks, one ion engine.
Lots of waiting.

The Mk.3 reaches orbit with about 10300m/s available Delta-V from the ion engine.

The launch process is pretty straight forward too -- fire the first stage more or less straight up as high as it will go, then circularize with the second stage at apoapsis.

Also interesting is that I've been flying these launches manually again rather than letting MechJeb do it for me. I've gotten to the point where I can get these things into orbit pretty much every time, and pretty much with an inclination of less than a couple of degrees.

This time I was more rigorous about the flight process, and wrote a log of the maneuvers and events encountered. I also changed the ion burn time to five minutes and used MechJeb for all of them. I threw away the launch upper stage after circularization -- everything after that has been done with the Ion engine.

Each Ion Burn was five minutes, and each time was 61.8m/s in Delta-V.

So I can say with authority that this was Ion Burn #14, that started at MET+2d8h44m, and lasted 1m20s:

Planning our escape.

This time the Munar periapsis was a more head-room friendly 124km.

Munar Periapsis, or thereabouts.
Kerbin rises for the last time.

And presto, we have escape at MET5d9h20m.  Doing the math from the recorded ion burns, I used up 760m/s to escape Kerbin from an initial orbit altitude of 197km.  We end up in good position, too, with the solar periapsis less than three days away.

So, having spent a couple hours of real time doing that, what should I do with my 9400m/s of available Delta-V?  I try some example nodes to estimate what the cost would be to go to Duna, Dres, and Jool.  

My back-of-the-envelope calculations have Duna costing 1500m/s; Dres costing 5000m/s (including the plane change); and Jool costing at least 5500m/s plus a plane change.  After considering these factors, I decide for Dres because:

• Everyone goes to Duna, it is a cliche -- a well done cliche, don't get me wrong, but I want to do something different.  And most people go there with Kerbals, me flying by with a probe isn't likely to impress anyone, not even me.
• Encountering Jool will be harder because of Jool's increased gravity and complex moon system.  Things happen faster in a high gravity field, and an ion engine is pretty slow.  Also we're getting out in the sticks again where the ion engine might not work well in the first place.  Put it all together and there is a good chance of a catastrophe.
• I can't remember anyone doing a mission to Dres.  So it will be new to me.

Dres it is!

Therefore, we'll do a Dres flyby and see what targets offer themselves afterwards.  If we do it right we might get a gravity assist to Jool or back to Eve.

And that's where I've left it for now.  I set up a 1h10m change-of-plane burn while I was writing this monster post up, and so I'll probably let that finish and then stop for today.

So, a couple things I've learned, in no particular order:

• The ionic burn point (periapsis, but can be anywhere on the circular initial orbit) should be either at orbital dusk or orbital mid-day.  Mine was orbital dawn, and dawn kept getting closer and closer to the burn time.  On ion burn 11, it was about 90 seconds before the burn started, rather close when you need the sun to burn.
• The MechJeb doofus seems to want to be placed on the right (starboard) side of the craft with the antenna pointed in the desired direction of travel.  Otherwise, the gimbal won't match reality.  (I thought I had a screenshot showing the gimbal level and the solar panels at 90 degrees, but I don't.)  My first try for the Mk.3 had the MechJeb on the top/front of the probe; but the gimbal was totally out of whack so I moved it back to the belly (opposite the radio dish), which is where it was on the Mk.2.  During one of the long periods between burns I played with the Mk.3's physical orientation to figure out where MechJeb really wanted to be.  We'll have to try that next time.
• When doing long burns, check in on the non-map view periodically to make sure your solar panels are still getting good sun exposure.  Otherwise, the engine might stop.

Oh, the change-of-plane burn finished.  The Hohman intercept burn will take almost two and a half hours to make the 1700m/s velocity change.  I'll set that up and let it run.

Review: Elysium

A little excessively violent and gory at times. A bit too much "District Nine" for me, but that is probably because we'd just watched that the night before. Worth watching, but not a DVD buy.

Review: District Nine

Slide between faux documentary and drama was jarring. High on the violence/gore spectrum for me. Interesting ideas, especially since they were not really explored.

Take A Toy To Work Day

In which The Cat In The Hat goes to work with Daddy.

(It is a G+ album because that's a little neater than what I can present here; each picture has a small message that I wanted Nathan to read to me so that he could share in the cat's adventure.)

M.A.D. Mary Keeps On Trucking

In which our final 0.20 mission is deemed a success.

Miracle On Minmus

I DON'T BELIEVE IT!!! I RESCUED IT!!!

(Context.)

Wiggling the steering started it slipping down the hill, and a little application of forward and back power -- and it gracefully tipped back on its wheels!

Miracle On Minmus

I suppose this is what the robot programs at NASA do -- fiddle with everything before abandoning the robot.

Well, now I have a functioning robot on Minmus!  I wonder how I'll lose it.

Amazing Amanda (previously Minmus-1j)
greets the dawn on Sol-1.

Closer, But Not Usefully So

So continuing on in the Minmus-1: A Chronicle Of Failure...

I spread Minmus-1h and -1i at high speed across a mountain side on Minmus. It is the same mountain, as a matter of fact, which gave me some handy visual reference distances when on final approach with Minmus-1j.

This landing was slightly more successful, although not usefully so:

Too much lateral movement and too much of a slope
makes Dave a sad boy. 

The title just about sums it up.  

I can spin the wheels but not do anything else with it. Maybe if I'd staged off the lander just before landing (and bounce) it would have either not rolled, or rolled one more quarter turn. Maybe if the rover had landed with its wheels facing in the direction of travel I could have caught it. Maybe... well.

Although it does occur to me that if I do actually succeed, I can "rescue" -1j by tipping it over with a more successful rover...

Kerbal and Mun rising in the background make it an interesting image, though.

And now I have more litter on the mountain side to act as a visual reference.  This is the debris field:

Hope there's no local littering ordinance.

Time for a change

Sometimes, change comes easier to me than I really expect it to.

About two months ago, I noticed that my watch was ten minutes slow.

Jokes about alien abduction and "missing time" aside, this disturbed me because I rely on my watch to tell me what time it is.  More so than the average person, I suspect, I am a little obsessive about knowing what time it is.  So suddenly being in a position where I didn't know what time it was was very disturbing.

I reset the watch and carried on.  A couple of weeks later, the watch was suddenly five minutes slow.

This really put me into  a situation because now I was wearing a device that I couldn't trust.

At work, I am the documentation chaser.  I am always on people to create and update documentation, for my use and for the rest of the team.  And one of my principles is that the only thing worse than no documentation is wrong documentation

Similarly, the wrong time is worse than not knowing what time it actually is.

So I stopped wearing my watch.

I've worn a watch pretty much nonstop since I was 12, so call that 30 years or so.  And I worried that I would not be able to make the transition to not having a timepiece on my wrist at all times.

It turns out that the time is usually close at hand even when I don't have a watch.  We are surrounded by clocks and appliances with clocks and computers and tablets and phones.  So I end up with only a very few daily instances when I don't have immediate access to the current time.

I have adjusted far better than I expected.

Maybe I am not so set in my ways as I thought I am.

Polar-ish Orbital Insertions

In which I accidentally discover a cool trick.

Minmus-1: A Chronicle Of Failure

In which repeated attempts to land on Minmus go hilariously wrong.

Eccentricity Valuation

I walked around all day with a screwdriver attached to my shirt collar.  It is a pencil-sized screwdriver with a shirt-pocket clip on it.  Today was clean up day this morning, so when doing the kitchen I clipped it to my shirt collar (because this shirt doesn't have a pocket) so I could take it back upstairs where it lives.

I'm not sure which is worse -- the fact that I walked around all day with it on, or that I'm considered eccentric enough that nobody I encountered (at the office or at a customer site) pointed it out to me.

Descent Into txtspk

And here's two photos of the advertising rotation on a major petroleum company's pumps:

Am I the only one who is bothered by this text-speak?  Is it really too much to expect complete, proper english?

And while I'm at it, those kids need to stay the hell off my lawn.

Reassurance

Somehow I am not comfortable with the idea that the people who wrote this website have written code that manipulates my hard drive at a low level:

Language matters.

The Problem

So I was slumming in the Linux Community on Google Plus, and this came up:

I presume as a joke.

And it occurred to me that this kind of attitude that Linux fan-bois have, this "you can build whatever you want" kind of attitude, is a problem.

But the big problem is that they don't understand that it is a problem.

Imagine if cars were really sold like this.  Where you bought a chassis and everything else,rig Dow to the trim and seat belt buckes, were individually optioned.  And then the new owner had to spend three weeks assembling the thing.  You would end up with a lot of cars built and assembled to a simple cook book put together by some authority.  And most of the resulting cars would be poorly put together, and probably quite unsafe.  Those built by the enthusiasts would be works of woder, one-off customs which excelled in some way or other, but on average most of us would drive pieces of junk.

Being able to buy a car that has been properly designed, and assembled by a professional workforce, lets us all drive cars which are, on average, much safer and better.  The cost of enthusiasts being bored is frankly an easy one to pay.

Actually it occurs to me that this is exactly the way kit cars are/were sold.  And the market spoke.  The sheeple wanted reliable, boring cars.  

But back to computing.  The fan-bois claim that you can do anything with Linux that you can with Windows; and while in broad strokes it is true (you can write a letter) it isn't in the way that most people care about (you can't write a letter in Word 2013).  

(But why do you need Word 2013?)

Because that is the format that all these other letters are in.

(So everyone should change!)

Yeah, good luck with that.

On-World Off-Roading

In which we play with cars.

Intentional Transit

In which Sally gets some company.

Impromptu Road Trip

In which an experimental orbital assembly accidentally ends up somewhere else.

Cunning Social Media Plan

Since we are, indirectly, on the topic of social media: if you recall, when I last mentioned this I was wondering how to participate in social media in a way that was good for both me and my audience.

My conclusion is that even though my audience is on Twitter and Facebook, I will probably move primary content generation to Google+. But I have a cunning plan to glue everything together.

See, I have located a service that generates RSS feeds from Google+ profiles.  I also have, in the past, written a perl script that can post to Twitter (which I have since lost somehow probably when the hard drive in the old Voyager died).  Twitter, as you know, can be read by Facebook.

Which means: wi the right perl script, I can post to Google+ and my audience elsewhere will either get the content, or a link to said content.  I will still monitor Facebook and Twitter, but for some reason the Google+ experience is making me happiest these days.

Now all I have to do is re-write that perl script.  How hard can that be?

Elementary and Sherlock

One of the series that we are watching this year is the Sherlock Holmes series Elementary.  It seemed like a suitable replacement for House, since House was basically Sherlock Holmes in a medical setting.  The main character feeds our need for an arrogant, antisocial ass who just happens to be able to draw useful conclusions from flimsy evidence.

Someone told me that Elementary was poor when compared to the BBC series Sherlock.  So since the pilot episode was one of the Apple freebies this holiday season, I downloaded it.  And yes, it was better, so far as it went.

We have purchased the rest of the Sherlock episodes available (all five more of them, as it happens) and are very slowly making our way through them.  Last night we watched S2E2 which means that we only have one more remaining unseen.

And that is where the principle difference between Elementary and Sherlock comes in to play.  Elementary is designed to be a weekly, 40-odd minute story with a beginning, middle, and end.  Sherlock episodes are more like self-contained movies that need to be spaced out much more because there is so much more in them.

Elementary is weekly TV, a weekly fix of entertainment; Sherlock is an event that needs to be scheduled and prepared for.

I read on Google+ (yes, I use Google+, stop laughing) that the second episode of series three has just finished principle photography, and now there will be a break in production to permit other commitments to be satisfied and that principle photography on S3E3 will start presently.  So within a year we should get the next batch.  But we will undoubtedly get many, many more Elementary episodes before then, weekly fixes that I look forward to.

Illuminating

In which we shed some light on things.

First Post

In which I publicly celebrate my heaviest lift to date.

Dave's Rules Of Home Renovation

Not that I'm an expert or been involved in many of these things,but it seems that one thing has been true every time we've done this:

1. Any renovation will take longer than you, or the trades, or anyone thinks.
2. ...even after accounting for rule 1.

Sharepoint Hate

The more I work with Sharepoint, the more I realize that I absolutely loath it.

It has the same total-lack-of-organizational-structure that a good wiki has, but throws in an awful web-based Word-style "formatting" and "style" markup toolset for presentation.  It has the same insistence-on-doing-the-wrong-thing that old versions of Word had (and maybe modern versions have if you have not come to grips with the formatting tools there).

The difference is, I can bully Word into A) staying out of my way until I want to do the formatting, and B) looking more or less the same as all the other documents I produce do.

The Sharepoint wiki thing is awful.

(I'm also not a fan of the Sharepoint ideal of just-write-everything-in-Word and storing those in Sharepoint.  The wiki is best in theory, but its like they've deliberately made it unusable.)

Somehow I elected myself policy master, and that means I get to write all the first drafts -- including the documents which describe how documents are to be written, named, and filed.  And being forced to use this tool just makes me want to hurt something.

The old wiki had the same lack of standardization and lack of ingrained organizational structure -- but at least I could do the markup language without thinking about it.

Since we are going to have to write another how-to-wiki document anyways, I'm wondering if changing our storage medium is really a good idea in the long run.  I'm starting to suspect that this is more about abandoning our messy past than actually fixing the future.

Dammit, Oracle

I don't want the Ask toolbar.  I've never wanted the Ask toolbar.  I've said no to the Ask toolbar each of the dozen or so times it has been offered in Java security updates.  Stop asking me about the Ask toolbar.

Algorythmically, how hard is this?

IF (<this is an upgrade of JAVA>)
 ANDIF (<the installed version of java shipped with the Ask toolbar option>)
 ANDIF (<the Ask toolbar isn't installed>)
   THEN ( the user probably doesn't want the Ask toolbar, so don't offer it )
ELSE
  IF (<the Ask toolbar isn't installed>)
    THEN(offer the Ask toolbar)

Seriously?

This is almost as bad as Adobe's insistence that I want a big, steaming pile of Adobe Reader icon plopped on my desktop during an upgrade of that.

I really wish I could just jettison java and be done with it.

Sooon....

Selinux

Major Hayden of Racker Hacker wants us to use Selinux. So he's created a site to shame us into using it.

My reply:

Seriously? Your simple solution to educating people is a 52 minute video? I don't have time to watch a video to figure out selinux. 

Why do I disable selinux? Because so many of the things I install include that instruction for setting things up. If the developers of these things won't take the time to learn how to make them work in a selinux environment, why should end users be expected to do this development work -- individually, meaning a lot of duplicated and inconsistent effort? 

It is like expecting end users to put in SQL input sanitizer routines in their web apps. You are targeting the wrong community. 

Yes, a system with SELinux is, all other things being equal, more secure than one without it. But a system with one-time passwords is more secure than a system without. Security is not just a question of "more is better". System administration is a case of balance, of trade off between utility, usability, and manageability.

If you can run in a selinux-enforcing worl, I have huge respect for you, because in the past I have tried and failed. 

But until I can install the tools I need/want and have the developers support me with selinux enabled, I have more important things to do.

Trolling The Twitter Advertisements: Mastercard

Standing Desk: 4th Iteration

I thought I had a picture of the third iteration of the standing desk, but I don't.

What happened was that a co-worker decided he needed a third screen because of what he is doing, so the boss got him one.  The boss asked me if I wanted one, so I jokingly said "sure" because I didn't expect it to happen.  But it showed up in January with other stuff for some new hires, and I'm not going to say no to new gear -- so presto, two screens.

I had them together on the cabinet on the desk, with two little Dell monitor stands underneath them, and this lifted them up to a height that was convenient for me to see.

The second thing that happened was that I lost my glasses.  This forced an emergency visit to the optometrist, and new prescription in hand, I went and got new ones.  So now, I can read the small print on the laptop LCD screen from where I stand.

So now I'm using things like this:

I'm not sure where the penguin came from.

I did have the keyboard on another two inch monitor stand briefly, but that got repurposed into a monitor stand for the two-monitor layout.  That left the keyboard a little uncomfortably high, and the mouse on the lower level.

Over the winter I went around in socks only for the most part, since the $10 shoes I was using were about $20 too expensive (if you know what I mean).  Since boot weather has more or less gone, I'm just wearing my shoes on the straight floor because the pad tile I have is too small to shuffle around on.  I am considering finding a larger mat to stand on, but with my day-to-day shoes I can stand straight on the floor.

I find I move around while standing, even if it is just shuffling back and forth on my feet while working.  And standing up makes it trivial to just step out of the cube when I need a micro break.

If I need to think, I'm still sitting down, especially in the afternoon.  This lets me stare out the window at the parking lot, which removes the distractions of the screens from my view.  There's also a white clear board I can doodle on in my line of sight, which helps.

But one good thing -- I happened to be on a scale last week, and I'm down 10 pounds from where I was at my last physical -- which predated the whole standing-desk thing.  So maybe it is paying off in terms of fitness, a little bit.

One thing is for sure, I've stuck with it a lot longer than I thought I would.

(Previously: 1, 2.)

Endless Circling

Dear Google,

When I click on Google+ in my web browser, the first thing you show me is a list of people I might want to put into circles.  Well guess what -- this is usually the same group of people you've shown me every time for the last six weeks (or however long this change has been active).

You are smart -- if I have not circled these people by now, what are the odds that I'm going to circle them now?  (Hint: it is the same shape as a circle.)

Stop showing me the same list of people.

Better yet, stop showing me a list of people.

If I want to circle people, I'll go looking for people to circle.  If I'm looking for Google+ , let me get to Google+ without the additional annoying click.

There are only two possible outcomes from you continuing this: either I'll just tune out the page completely and automatically click the "continue" button, or worse, I'll eventually get so mad I'll go ARGH! FUCK! and just stop coming here.

And you wouldn't want that, would you?

Didn't think so.

Love, me.

Dilbert On Being The Firewall Guy

It is rough being the firewall guy.  People always blame the firewall, probably because A) it is a box which by design prevents things from happening, and B) the average end user has no control over.  So as a firewall guy you end up fixing a lot of completely non-firewall-related issues because that's the only way to prove that the firewall isn't at fault.

And yet, some of us voluntarily choose to be the firewall guy.  Possibly because you can "accidentally" prevent people who are mean to you from getting at the internet.  Not that ever happens, no, people tend to be nice to this firewall guy.

Now, anyways.

Things I Learned (a continuing series)

The animators responsible for the Dire Straits video Money For Nothing, Ian Pearson and Gavin Blair, went on to found Mainframe Entertainment, who produced other ground breaking works such as Reboot! and Shadow Raiders.

Being Sad: Trying A Couple RSS Options

Dealing with being sad in a post-Google Reader world has led me to try a couple of RSS feed readers.

When looking around, I automatically excluded readers which depended on the Google Reader back-end to do their data storage and feed management.  I may try them after Google Reader is shut down, but right now there's no point in finding something I like only to be let down again when they don't update to another back end (either immediately or, as is the case for some iOS apps, ever).

On the iPad, I've been using Feedly.  Feedly uses the Google Reader back-end, which yes should disqualify it.  But they claim to be building their own back-end to replace it so that the transition will be invisible to end users.  Initially I was turned off by its magazine-style flipboard which shows only a few articles per page; however I've played with the options, and a recent rebuild of the app added a mode which seems to display more stories per "page".

Feedly is painful to use in a desktop browser, though.  I'm highly suspicious of these Chrome plug-in things, and Feedly wants to be one of those -- I suspect it is trying to manage a local cache of all my feeds rather than having me hit a central server for them.  (Which makes me wonder how the iPad app actually works.)  The real killer, though, is that Feedly seems to be littered with duplicate articles.  The sorting seems haphazard, and there just isn't the same article flow that was in Reader.

So for the browser, I've set up an instance of Tiny Tiny RSS on my private server.  And for a private instance it isn't too bad.  And as a side-effect, I've pruned my RSS subscription list down from more than 300 to just a hair over 200 -- a few broken feeds that I decided I don't care about, feeds with no updates since 2011 (or earlier in some cases), or just plain 404-ing feeds.  It works OK in the browser, and I've come to grips with the configuration options that are presented.

It'll do a number on the bandwidth usage that my server uses, though.  We'll just have to wait a couple weeks to see how running the updater twice an hour will impact my data transfer numbers.

But on the iPad, the mobile interface is just plain painful.  Or at least, on my iPad.  There's a ton of hesitation and reloading and spit-you-back-to-the-default-page-load that makes it almost unusable.  And since a non-trivial percentage of my feed reading is done on the iPad, this is a problem.  And running it in the iPad Chrome browser is an order of magnitude worse than that.

Oh, I'm also a bit put off by the TTR dev's attitudes which seems to be "we're not changing TTR to be like Google Reader, so piss off".  Which is fine if you like where the boat is going.  And normally I would agree with them, it is just their attitude is lacking.  I do think that expecting end-users to do their updates with git, claiming "the trunk is usually stable and usable," smacks a little bit of not caring about how others see your tool.  The install is the first time a lot of people see your tool, if the install is rough they are more likely to give up on it and be less tolerant of other rough edges.

Downsides: I thought that I could deal with seeing articles in both TTR and Feedly, but I'm less happy with that than I thought.  And there's the whole problem of managing RSS subscriptions in two places -- there's no trivial way for me to duplicate my feed trimming in Feedly from TTR.

I've been trying hard to not go back to Google Reader, since that won't be an option in the future.

I'm not sure what I'll end up doing in the long term.  But I'm not exactly thrilled with how things are going right now.

Anti-Social Media

Continuing on our theme of internet corporations making me sad,  it has come to my attention that the very thing which I lauded twitter and facebook for is about to stop working: at some point in the near future, Twitter will break the Twitter-to-Facebook integration that I rely on to keep entertaining my public.  This is because Twitter has decided that they are no longer just a pilot fish in the great internet sea -- they are a destination that is important in and of itself, and therefore there's nothing to be gained by letting Twitter content leak out to other social platforms where someone else can monetize it.  If you want The Twitter, you can come to The Twitter, and The Twitter will happily serve you ads along with the content you desire.

For those of us further up the product foodchain, though, the question becomes: now what?

There is a temptation to just simply shift the burden to the reader and force them to charge around to the various social media sites to pick up the pieces of my communications that they care to receive.  This is by far the easiest for me, as it means I can just write what I want, when I want, and where it occurs to me, and my faithful readers will of course spring to action and visit all of these lovely sites.

Yeah.

A scenario made even more unlikely due to the fact that A) none of these services let information out in RSS format and B) since Google Reader is being retired, there isn't a decent RSS reader left to read these non-existent feeds anyways.

I'm still seeking social media nirvana, where I can post something and it is automatically disseminated across all the services I use.  This will never happen, since Facebook and Google are sworn enemeis, and now Twitter flatters themselves that they can sit at the same table.

Let's briefly look at the competition:

Twitter: microblogging.  Not suitable for complex thought, but it is pretty interactive.  The advertising is pretty unobtrusive (and currently totally absent if you use a client like Tweetdeck, although rumor has it that Tweetdeck is also going to be binned).

Google Plus: Google just loves you for your data, your sweet sweet demographics data that lets them serve advertising somewhere to.. someone.  No one is really sure.

Facebook: all the privacy-violating goodness of Google, except the someone they are serving ads to is you and your readers, and the ads themselves are hilariously bad.  Oh, and the chance that there's an actual audience on Facebook is only about 500% higher.

Audience-wise: I think it is a toss up between Twitter and Facebook for numbers.

I've toyed with a number of ideas: there's a service which allegedly will post from one social media platform to another, all you have to do is give it your credentials for everything (ick).  Or maybe just doing everything in Blogger (hey!  a fourth platform!) and trying to find some like-on-Facebook, tweet-with-Twitter, and +1 buttons and just pushing all three myself when I post something here.  That'll get the word out, and drive traffic back to me here.  Of course that just turns this Blogger thing into a Tumblr (hey!  a fifth service!), doesn't it, so maybe I should just go there and put the buttons on posts there?

I think I still have a tumblr... I wonder if I wrote down the access credentials anywhere.

Anyways this is all forcing me to make decisions, and everyone knows how much I love decisions.  So nobody is holding their breath on this.

I'm sure there will be a general announcement if and when I decide what I'm doing.

A Defense Of Jargon

Rhett Allain calls for science to stop using three words: hypothesis, theory, and law.  The crux of his argument for removing these words is because the common people think they know what those words mean, and the common conceptions of those meanings is inconsistent with science's use of them.

He's wrong -- and in fact, I think his arguments are potent when arguing for more jargon in science rather than less.

Mr. Allain's issues problems with modern language are as follows:

• The real world is very complex
• There are only so many words
• Therefore, context is important; and implied context king.

Let me give you an example.  Consider the word "bridge".  Now to a mechanical engineer, dentist, and computer scientist, that word means three completely different things.  Yet if you asked the common person what a "bridge" was, they'd go with the device to permit transport of something over some other obstacle.

There's no context embedded in the word, so you have to waste more imprecise words to build a context around the word which conveys the meaning that you wish to convey.  And in today's world of the five second soundbite, there's no time to build that context.

The legal profession is brutal in this respect, because the vast majority of the words and grammatical constructions that they use in their communications are identical to those found in common English   The problem is that in the legal world those words and grammatical constructs have precise meanings without the ambiguity that common English has.  But a legal commoner like me can read a legal document, understand all of the words, and come out at the end with a completely different meaning that a lawyer would.

The medical profession gets this right in a lot of ways, because they draw from latin words to describe precise parts of the body.  There's no ambiguity about what "thyroid" means or what procedure a "phlebotomy" is.  The words have single meanings in common use, which means they have their own context more or less embedded in them.  And better: since this context is embedded in the words, it acts as a signal to the common user that there is specific knowledge required that they may or may not have.

There is a collision here, between the media-fueled short attention span, and the increasingly complex concepts in the wider world that can't be fit into shorter and short sound bites.  Inevitably people will get left behind in certain areas -- I may be well versed in systems networking, but I am incapable of reading a legal document, or understanding in depths the medical subtleties that arise from specific treatment options.  I am forced to accept my specializations and follow advice from lawyers and doctors, or put aside my expertise and begin studying these other areas on my own*.

I think that rather than trying to soft-pedal this complexity in the modern world, we need to be shoving it more up the noses of the uninformed, telling them that they should either educate themselves and participate as a proper member of the community, or get out of the way of those who are.

And bending language to being more precise by embedding context into soundbites is a good first step.

---

*: And I think this is the difference between accepting advice from a scientist, doctor, or lawyer, as compared to a spiritual leader.  With time and effort, I could educate myself and become a scientist, doctor, or lawyer. That knowledge can be freely gained, all you need is time and effort. (Complaining that you don't have either of those does nothing to change this fact.) The problem with spiritual leadership is that there is no impartial mechanism you can use to come to same conclusions.  You can't test the first principles for yourself.