Skip to content

Function heirarchies and getting the computer to do what you want it to do

It appears that Shuttleworth has noticed a lot of the hate about Unity. He also seems to have a good understanding of why it exists and he has a plan.

We’ll resurrect the (boring) old ways of displaying the menu in 12.04, in the app and in the panel. In the past few releases of Ubuntu, we’ve actively diminished the visual presence of menus in anticipation of this landing. That proved controversial. In our defence, in user testing, every user finds the menu in the panel, every time, and it’s obviously a cleaner presentation of the interface. But hiding the menu before we had the replacement was overly aggressive. If the HUD lands in 12.04 LTS, we hope you’ll find yourself using the menu less and less, and be glad to have it hidden when you are not using it. You’ll definitely have that option, alongside more traditional menu styles.

“Overly aggressive” in making changes? That is probably the best way to put it.

The plan is about HUD – head-up display.

This is the HUD. It’s a way for you to express your intent and have the application respond appropriately. We think of it as “beyond interface”, it’s the “intenterface”. This concept of “intent-driven interface” has been a primary theme of our work in the Unity shell, with dash search as a first class experience pioneered in Unity. Now we are bringing the same vision to the application, in a way which is completely compatible with existing applications and menus.

The basic idea is that of a look-ahead search through the menu tree that is context based and inferential. You may not know exactly what you want or how to spell it but you can guess and see what is in the menu tree that might fit. Instead of walking through the menu branches taking guesses about which one might have what you want, you can have the system help you find it.

Shuttleworth is not working in a vacuum and he is aware of the other efforts on the same problem.

There are other teams interested in a similar problem space. Perhaps the best-known new alternative to the traditional menu is Microsoft’s Ribbon. Introduced first as part of a series of changes called Fluent UX in Office, the ribbon is now making its way to a wider set of Windows components and applications.

He notes that the ‘M’ in WIMP is for menu and refers to the menu idea having a 30 year history with no revolution. Back when the functionality of programs was rather small, grouping them into a hierarchy of the 5 to 7 things a human could have in mind at one time wasn’t a hassle. These days, that 5-7 comprehension idea is getting stretched so far that the complexity and capability of modern applications is getting cumbersome. We live in interesting times and it appears we will be seeing a revolution in how we access the capabilities of our computers and applications.

2012 01 24 bryan Innovation and entrepreneurship
Ubuntu and its ilk
we live in interesting times Comments (0) Permalink

Getting around the burden voltage in measuring low currents

Current is usually measured as a voltage drop across a known resistance. David L. Jones notes that this ‘burden voltage’ can be a significant disturbance in low voltage and low current circuits. His solution is the µCurrent. He provides a description and theory of operation as well as a circuit diagram for the device at his EEVBlog.

There are three parts of the circuit. The coin cell battery status monitor LED uses a ‘power supply supervisor’ chip that makes it easy to check the lithium coin cell to make sure it is providing the needed voltage. Another part of the circuit splits the battery voltage in order to provide the reference op amp with the supply voltages it requires. The reference op amp simply converts the signal from the dropping resistor into a voltage that can be read on a DVM in a convenient way to determine current.

I guess that these days the input impedance of voltmeters isn’t that much of an issue. It was that long ago (it seems to me) that a 20k ohm per volt meter was something significant. Vacuum Tube Volt meters were the cat’s meow when it came to measuring voltages and current. That was when you needed a vacuum tube to get a high impedance amplifier to isolate the input and get enough output to drive a D’Arsonval meter.

an aside: I am in the process of redoing the post categories and immediately after a first pass to try to make sense of things, along comes another interesting topic that doesn’t quite fit. Where do I stash a ‘build your own’ and ‘circuit theory’ and ‘measurement theory’ ?? I’ll have to think about this. There are tags for individual posts so I’ll have to use them until I can figure out a better solution.

Tagged , ,
2012 01 21 bryan Education and Support
Hardware Comments (0) Permalink

Words mean things?

Cognitive scientists develop new take on old problem: why human language has so many words with multiple meanings at PhysOrg has some interesting implications for legal and technical prose as well as for the nature of some arguments in the online forums and discussions.

To understand why ambiguity makes a language more efficient rather than less so, think about the competing desires of the speaker and the listener. The speaker is interested in conveying as much as possible with the fewest possible words, while the listener is aiming to get a complete and specific understanding of what the speaker is trying to say. But as the researchers write, it is “cognitively cheaper” to have the listener infer certain things from the context than to have the speaker spend time on longer and more complicated utterances. The result is a system that skews toward ambiguity, reusing the “easiest” words. Once context is considered, it’s clear that “ambiguity is actually something you would want in the communication system,” Piantadosi says.

The difficulty arises when the speaker and listener do not have the same goal. The listener uses ambiguity to distort and twist the meaning the speaker is trying to convey. That forces the speaker to devolve into excessive explanation and definition of concepts to reduce the ambiguity.

The purpose of the study probably isn’t to understand flame wars on the I’net. It is more likely aimed at helping those working in natural language processing (NLP) so devices can respond to spoken commands more like humans do.

2012 01 19 bryan factoids
social Comments (0) Permalink

Viewing distances

Ecoustics has information on optimal HDTV size and viewing distances. The formulas and tables are based on the work of Bernard J Lechner who was an EE and VP at RCA. For 1080p, the optimal viewing distance works out to be about 1.6 times the diagonal screen size. Since the HDTV is becoming rather common as a computer monitor, the viewing distances have information for that use as well.

The thing to keep in mind is that you really don’t want to be on the ragged edge of seeing every last pixel in your display. There is also the issue of viewing angle with the wide screen displays. As a computer monitor, being close for extreme detail and having the screen occupy a wide swath of your vision real estate might help productivity. For media viewing, the opposite might be better. So, a 32″ HDTV calculated at best at a 4′ viewing distance might be good for computing from a distance of 3′ and for media viewing at 6′ –

Rules of thumb can be useful but keep in mind that they aren’t specifications!

2012 01 18 bryan Multimedia Comments (0) Permalink

Final judgment on the doornail: Dead as it appears

Dr. Dubik and Dr. Wood thought it necessary to make sure that ‘dead as a doornail’ was indeed an appropriate analogy. Since advanced life support technology has confused things somewhat and there are now laws concerning brain death, standards matter. So How Dead Is a Doornail?

If the definition of death as expressed by the AMA et al has validity, it should be possible to compare this recent criteria against the widely accepted and time-tested “doornail” standard. We did just that.

We subjected a large doornail (see Figure 1) that was forged in 1986 to thorough examination, prolonged close observation, and an electroencephalogram (EEG).

Their studies resulted in 4 findings that conclude “that the criteria for death as described in modem medical literature 1,2,3.4 is valid and may be used with confidence by clinicians.”

Caution: the post contains pictures of the subject under examination that may be disturbing.

OK. so now we know. That door nail was indeed without life. Or at least the one the doctors subjected to examination met the criteria. Maybe other nails?

2012 01 17 bryan Education and Support
factoids
we live in interesting times Comments (0) Permalink

Building things in the early days

Bootstrapping is not a trivial task. A Spellchecker Used to Be a Major Feat of Software Engineering is a reminder about the bootstrapping phase of software development for personal computing.

Here’s the situation: it’s 1984, and you’re assigned to write the spellchecker for a new MS-DOS word processor. Some users, but not many, will have 640K of memory in their PCs. You need to support systems with as little as 256K. That a quarter megabyte to contain the word processor, the document being edited, and the memory needed by the operating system. Oh, and the spellchecker.
…
Fast forward to today. A program to load /usr/share/dict/words into a hash table is 3-5 lines of Perl or Python, depending on how terse you mind being. Looking up a word in this hash table dictionary is a trivial expression, one built into the language. And that’s it. Sure, you could come up with some ways to decrease the load time or reduce the memory footprint, but that’s icing and likely won’t be needed. The basic implementation is so mindlessly trivial that it could be an exercise for the reader in an early chapter of any Python tutorial.

When I was going to school, the IBM 360 was the big machine. That had main storage often ran about a megabyte or less. That was when memory was little magnetic cores with lots of wires threaded through them. It always struck me as funny that this computer was built on the idea of 4 letter words – it was a 32 bit machine with byte addressable memory. It was the transition machine from discrete components to the integrated circuit. This is also the time when minicomputers, such as the PDP 11, showed up and Unix appeared. There was a significant difference between the business computing hardware and the academic stuff.

It was at the tail end of the IBM 360 product life that the PC showed up. That was an 8 bit PC with 64k of addressable memory and a couple hundred K available on floppy disk. That was the PC for VisiCalc and Electric Pencil. When the IBM PC came out with a 16 bit processor and 1 MB of addressable memory, the spreadsheet and word processor really started to stretch out.

One of the problems those of us who learned to write programs back then have today is to get a handle on the wealth of today’s computing environment. No, we don’t need to be anywhere near as careful with memory usage and processor limitations. These days, the languages for software development have constructions that handle many of the housekeeping chores that kept us busy in the past and there are libraries that encapsulate just about anything that has been done in the past.

Spell checkers make for a good example. At first, they were simply of the ‘find a match’ variety. Then they added the ability to guess at what you meant so they could suggest a correction. Then they became background processes that could help you as you typed in text. Contrast to grammar checkers which have a much more difficult task – or maybe facial recognition which is showing up in modern point and shoot cameras that recognize your family members to set priority on its automatic focus.

There are times when we should take a look at the innovation and invention that we take for granted. We live in an amazing time of growth in the intellectual realm.

2012 01 16 bryan Applications
Innovation and entrepreneurship
Tools
we live in interesting times Comments (0) Permalink

1 to 10 kWh battery systems – rules of thumb

You’d think the Amateur Radio types would have Ohm’s law in their toolbelt and not be afraid to use it. Between QRZ forums and this month’s QST, it doesn’t seem so when it comes to emergency backup power systems. For medium sized solar systems with battery backup, systems in the one to ten kilowatt hour range where the battery of choice is the lead acid type commonly found at retail for cars and RV’s, there is a lot of ignorance, myth mongering, and presumption of ignorance. Here are a few rules of thumb I offered to a query on the topic.

  • Figure about a usable 15 watt hours per pound of lead acid battery for available energy capacity. This is the 50% DoD target often used as most cost effective.
  • Figure at least 1 watt of solar panel per pound of battery.
  • Plan on an energy harvest from solar panels equal to the maximum rated output for 2 hours (for 240 watt panels plan on 480 watt hours per day from the solar). You might get a factor of 3 to 5 instead of 2, maybe, but don’t plan on it.
  • Figure you’ll need 4 watts per pound of battery for a proper bulk charge. You need this recharge power at least once a week for batteries in use daily.
  • Figure any battery energy capacity measure has a 10% to 20% variability due to factors such as age, cycle to cycle variation, temperature, and use profile.
  • Note that the 20 hour energy capacity rating is at about a watt per pound of battery. Draws higher than this will result in lower available capacity (re Peukert).
  • Don’t let a 12v lead acid battery go below 12.0v as measured after it has had no significant charging or discharging for at least a half hour. (and that’s pushing it. recharge at 12.4v and keep above 12.2v if you can).
  • Out of use storage maintenance for batteries needs a device that will maintain a full charge without overcharging and will apply a sulfation inhibiting technique. Float alone just doesn’t cut it anymore. (if you have electrolyte loss or corrosion at the terminals or poor charge retention, storage maintenance charge is a good first suspect)
  • You will need a decent fuse at the battery. These are usually 100 or 200 amp fuses. This can then feed a main line to your operating table which will have a fused distribution center feeding fused power leads to the equipment (that’s three fuses between the equipment and the battery).
  • If you plan to run more than a couple of hundred watts, especially with large surges (like, say, using an inverter or cw), then then impedance of the battery feed can be something to consider.
  • Watch out for RF noise from solar charge controllers, especially MPPT types that you need so you can run higher voltage solar panels with smaller resistance loss in the wiring.
  • You may want to use a boost controller so you can feed a solid 13.8v to your rigs from battery power.
  • While you don’t need to go overboard on wire size, a bit of overkill here, especially for any run more than a couple of meters, might be worthwhile. Figure a millivolt per foot per amp for 10g wire and a resistance that differs by a factor of 2 for each 3 wire gauge size change. Try to keep voltage loss in the line below 5% for maximum loads and keep in mind that wire length is the distance out plus the distance back.
  • Don’t use frame ground as a power return lead but do watch out for ground loops.
  • Crimp connections with proper fittings and a good tool. Don’t solder. Use silicon dielectric in the fittings to help reduce corrosion.
  • If you don’t like what you get, try something different. Learn from your experience and use that learning.

There are indeed a lot of ideas and condensed knowledge in this pile. The Solar Calculator : Weather Underground will calculate maximum solar energy harvest and even provides for the particularities of a number of common solar panels. A lot of the battery stuff requires critical reading and getting past a lot of the bullhooey (such as in the QST article about ‘deep cycle battery’) to generalize what can actually be measured and useful.

2012 01 13 bryan Amateur Radio
Education and Support
Off Grid Comments (0) Permalink

Missing the point; Edison vs Tesla; AC vs DC power

DC for data centers at Gigaom has some good history and a few interesting tidbits on a trend towards DC in data centers. They miss the big issue, though.

The reason that Tesla and Westinghouse won out over Edison in establishing AC as the power distribution standard was about electrical power delivery over distance. With AC, transformers made it easy to step voltage up for transmission lines. That reduced current needed for a given power level and that reduced current reduced losses. There were other advantages, especially back in the day, as equipment power supplies and motors could also be made more efficient with AC (remember vibrating power supplies for car radios back when 6v and tube radios were the thing?) (See wikipedia on power transmission)

As Kanellos notes at Gigaom, DC is gaining some popularity. Part of this is the ‘alternative’ energy systems that depend upon batteries, part is the local distribution of power, and part is technology advances that make conversion between AC and DC less expensive and more efficient. These advances have also made power transmission of DC more economically feasible in some cases (see the Wikipedia article)

To solve the mismatch, a whole industry of AC-DC converters has been developed. National Semiconductor sells billions of dollars of worth of chips to convert power. Inverters in the solar industry exist to converts DC from solar panels to AC that can run on the wires in your home.

The data centers tend to use medium voltage DC – 48 volts. Low voltage – 12 VDC – can be found for most of the wall warts you find with external disk drives, small TV’s, and other such equipment.

AC power transmission has a lot of interesting engineering when in a grid. Keeping the phase in synchronization between all parts of the system can be a challenge. Impedance matching with loads can be a challenge. Noise, like lightning, can be a problem. The fact that we depend upon electrical power as a matter of course and rarely even notice its presence says a lot about an amazing engineering feat.

2012 01 13 bryan Education and Support
Hardware
Off Grid Comments (0) Permalink

Digital communications methods

KB6NU noted Yaesu’s A Digital Communications Guide for Amateur Radio Operators as a good introduction into digital radio communications.

The ‘white paper’ does provide a good rundown on the benefits and drawbacks of digital communications. Many have seen those first hand when TV changed from analog to digital and they were trying to get off-air broadcast signals for their TV.

The paper does promote Yaesu’s four level FSK digital mode as a contrast to Icom’s D-STAR but, to me, that is more of a compare and contrast rather than a bash and trash.

What is in the paper is that the traditional analog systems that Amateur Radio operators have always used have their place and will likely continue to be popular. Digital modes will remain mostly experimental and of limited utility in the hobby.

The paper features the three levels involved in digital communications. There is the modulation type like GMSK or C4FM, the communications type such as frequency or time domain multiple access, and the protocol that is how data is encapsulated for transmission. In analog systems, the modulation type is the only concern.

The fact is that digital modes have long been a part of Amateur Radio. RTTY was one of the first (after CW). Packet showed up something like forty years ago. Since then, there have been so many digital modes that keeping track of which one you are hearing can be a job. That was why software like fldigi was developed. Then there is the weak signal digital effort exemplified by WSJT.

Note that this discussion of digital is about communicating digital information. It is not about digitizing radio signals so they can be mathematically processed. That distinction is between digital outside of the radio and digital inside it. They are two entirely different technologies.

2012 01 08 bryan Amateur Radio Comments (0) Permalink

LEDifying a Ten Tec Triton IV – part I

I dug out the old Ten Tec Triton IV (now called a model 540 as Motorola complained about using its trademark) transceiver so Dave, KA7VLL, and I could play with it. First up was to replace the panel lamps with LED’s. Under the idea that ‘when you get a new hammer everything looks like a nail’ this was an exercise with LED light strips – see LED lighting – DIY strip lights. Here are some pictures to show what we’ve done so far.

making holes for light to shine through

strip trimming

The LED light strip, at 8mm, was a bit too wide

A modified panel and a modified LED strip together to check for fit

One module of an LED light strip for the S meter

The LED light strip was a 5m long circuit board with 100 circuits. Each circuit had 3 LED’s and a limiting resistor in series for a 12v nominal supply. It appears that they will function with the existing series dropping resistor that was used to dim the original bulbs a bit. That additional dropping resistor is needed to keep the LED’s from blinding the operator.

Wire wrap wire was used to connect the LED’s together to the power access points the original lights used.

What’s next? Before we can really see these LED’s in all their glory, Dave and I will replace the dial string and lube up the PTO.

For the dial string, I did have a shock cord rescued from a box of candy for Christmas to replace the 35 year old stretched out one in the rig. It was a bit bigger diameter but looked like it’d stay in the pulleys and do the job. The problem was that the .020 Dacron string was a tad short. (You can see the string wrapped around a couple of pulleys and going over the S meter in the picture of the S meter LED’s). That was why the adjustment cam was set to minimal effect. So a new, calibrated, part of the dial string assembly was needed.

Dave sprayed some contact cleaner into the PTO bearing and that did loosen things up a bit but the gears were not really happy. It turns out that Ten Tec sells both dial string and PTO lube kits. The dial string kit cost has increased at about twice the inflation rate so the 25 cent cost mentioned in the manual is now a dollar something. The PTO lube kit is supposed to have a good guide to the recipe for getting to the shaft bearings as well as the proper lube and whatnot. It’s cost was $35 and should get the main tuning knob back to ‘as new’ condition if we do it right.

The Triton IV was one the first all solid state, no tune, amateur bands only, 100 watt class, transceivers. It was known for its CW full break in performance. It’s also a good radio for learning about how these things are built. The block diagram has a circuit board for each block so you can hit the hierarchy of functions at the top level and work your way down to individual circuits and then to the components used to implement that circuit. Lot’s of room, mostly discrete components, and well laid out to see what it does.

The value of these things at flea markets and online appears to be about $250 or so. This one has been through a few ‘learning experiences’ so it isn’t in an OEM state. That is one reason I was able to handle it when Dave grabbed the Dremmel to start cutting a few slots for the LED lights ….

2012 01 07 bryan Amateur Radio
Hardware Comments (0) Permalink