ISOTYPE Book: Mackenzie, The Vital Flame

By | ai, bigdata, machinelearning | No Comments

(This article was originally published at eagereyes, and syndicated at StatsBlogs.)

The first book in the new series on ISOTYPE books is The Vital Flame by Compton Mackenzie, published by The British Gas Council in 1947. It contains 42 color photographs and five ISOTYPE charts, with a nice variety of different topics and styles.

A Bit of Context

My copy of the book came with a letter from the County Clerk of Lanarkshire to a Mr. Law, sent in May 1947, presenting him with the book. The letterhead gives a great idea of the time: it lists the telephone number as “Bell No. 3371” and an address for telegrams.

The letter is a form letter sent to a number of people. It looks like it was copied (perhaps with a mimeograph or a spirit duplicator), with the name of the addressee filled in with a typewriter later. It tells Mr. Law that the council agreed to send out “a number of copies” of the book to people likely interested in this topic. It’s rather vague and mostly pointless, telling the recipient that the council has received the books and so is now sending them out. Well, then.

The Book

The inside title of the book nicely summarizes the book’s “forty-two pictures in Colour Photography” and “seven pages of Isotype Charts.” For some reason, the capitalization Isotype seems to be very common in many of the books I have seen. Strangely, there is no table of contents for the 80 pages of text, but there is a List of Colour Plates for the photographs and a List of Charts in Colour for the ISOTYPE diagrams.

The Vital Flame is one of those books that they just don’t make anymore: a no-nonsense description of how gas is made and used for different purposes. There are little stories, but for the most part the writing is very straight-forward explanation, sometimes getting more technical. There are explanations of technical topics like the chemical formula and properties of different kinds of gas, how to calculate therms and BTUs, etc. There’s also a bit of history of gas used for lighting and heat, and how that caused all sorts of interesting problems with different industries, such as coal (which felt they were being replaced), and even the Royal Navy (which apparently worried that with less demand for oil from whales, there would be fewer people trained as crew on ships, and thus potential hires for the navy). Mackenzie is quite opinionated in the history part in particular, which is fun to read. The latter parts get dryer, except for the stories from the war.

The photographs give lots of interesting insights into the mid-1940s. This one is captioned “No more moths. D.D.T., used against typhus during the war, gives the housewife a new insecticide.” After being widely used for decades, DDT was banned in the 1970s for causing cancer and killing birds and other wildlife in addition to the pests. The role of women was about to change dramatically after their involvement in all kinds of jobs during World War II, but 1947 hadn’t quite gotten the memo yet.

Some of the technology is also a good reminder of where things were just after World War II, such as this oven for baking what looks like a classic British meat pie.


As far as I can tell, the text never references the charts, and certainly never explains them. They are included at the very end, unlike the images that are interspersed throughout (though on separate pages and different paper). Interestingly, the book was printed in three different places: one printed the text, one the color photographs (on glossy, heavier paper), and one printed the five-color ISOTYPE charts at the end (on slightly heavier but non-gloss paper).

Two of the charts are printed across both pages, and three more are single-page. They all bear the logo of the ISOTYPE Institute.

The first chart compares gas production in classic ISOTYPE fashion between the U.S., Great Britain, and a few more countries. The book talks about manufactured gas, which was apparently the only type of gas in use at that time in Europe, as opposed to natural gas (common in the U.S., according to the chart).

This chart has all the makings of ISOTYPE: repeated symbols that represent a multiple (500 cubic feet per person – and year, presumably), a title that says just enough to read the chart as a comparison, and a small legend explaining the numbers. The two categories are labeled directly on the chart.

The next chart is quite different, because it doesn’t show quantities. Instead, it depicts the process of coal carbonization and then lists the different uses for all the different products that come out of that. The colors represent different kinds of uses: orange for fuel, red for indirect or industrial uses, green for agriculture, and blue for direct consumer uses. The little symbols, even stylized as they are, show the time quite nicely, like the fuel pump at the top, or the perfume bottle in the lower right.

Back to classic ISOTYPE, we see annual gas sales in Great Britain from 1885 to 1945. The time periods are of different lengths, but since the numbers are per year, that should not matter. It’s a classic ISOTYPE chart, with some embellishments showing the evolution of lighting, heating, and cooking, from the late 1800s to what was then a state-of-the-art kitchen oven.

The next chart is the rather unusual ISOTYPE map, showing gas consumers by region. Each house represents 50,000 families, and the houses are grouped in 5s to be easier to count. What confused me initially is that the legend only explains what blue means (gas consumers), but not black. I assume black simply means that those households do not have gas in their homes, but it would have been nice to spell that out. I like the little drawing at the top, showing the gas flowing from the gasometer to the houses.

Gas consumers are clearly the vast majority overall, the only region with more non-consumers than consumers is West. The comparison is a bit difficult because the numbers of households are very different between the regions. But given the vast differences, that doesn’t appear to be a big problem.

The final chart shows a bit of a twist on ISOTYPE. The flames represent the amount of gas and coke used in terms of BTUs, compared to the amount of soot deposits shown as black “clouds.” Clearly, as the amount of gas consumed increases (at the expense of directly burning coal, presumably), pollution gets better. Though 200 tons of soot deposits per square mile still strikes me as eye-poppingly terrible. I don’t think we can even imagine today the kind of air pollution that was prevalent in London through the first half of the 20th century.

From what I have seen so far, The Vital Flame strikes me as a typical ISOTYPE-containing book from around the middle of the last century. It aims to inform the reader about a topic, but mostly stays away from numbers in the text (there are few here and there). The ISOTYPE charts are separate and – I’m guessing – were commissioned and created after the text had been written. Other than on the title page and on in the list of charts, they are never mentioned. Other books advertise them on their covers, though this one doesn’t. Including them must have come at significant cost though, not just to pay for the charts to be made, but also for the more complicated printing process in color.

Given the photographs (in color! quite a remarkable for 1947), the book didn’t need the charts to be less dry or more approachable. I wonder if they were meant to make the book appear more authoritative and data-based.

Please comment on the article here: eagereyes

The post ISOTYPE Book: Mackenzie, The Vital Flame appeared first on All About Statistics.

Source link

Smraza 4 in 1 Starter Kit for Raspberry Pi 3 2 with Clear Case,Power Supply,2pcs Heatsinks and Micro USB with On/Off Switch

By | iot, machinelearning | No Comments

The best combination and price, it is including all your needs for your Pi 3B and RPi 2B. You don’t have to run around trying to find these accessories in different places. The case is specially designed for Raspberry Pi 3B 2B, access to all controls and features.Note: (Raspberry Pi Board and SD Card are NOT included)

Clear Case: All the access ports are accessible, SD card can remove easliy when the RPi install in the case. And the top panel is designed for easy access to the GPIO pins and easy to plug and remove anything so that you can easily to connect it.
High quality material: This case is made from high quality PC material, which has the advantages of good high temperature resistance, impact resistance. And you can put much pressure or weight on the case without danger of cracking.
Good heat radiation: The case design has enough space to add heat sinks, and the bottom ventilation design allows sufficient air circulation.
USB Cable with ON/OFF Switch : It is convenient for you to turn on and off your RPi.(Not for data transfer use)
Power adapter: 5V 2.5A Power adapter:Full Output 5V DC 2.5A Regulated Input 100V to 240V AC.

There has 3 heatsinks in this kit and two heatsinks(A large one and a small one) are sticking on the front of the board.One large heat sink(10mm tall) for CPU and one small heat sink(5mm tall) for the LAN ,the last small heatsink is as standby.

Package Included:
1) Clear case x1
2) heatsinks x3
3) 5V/2.5A Power supply adapter x1
4) Micro USB cable length:39.5inch/(1meter) x1

About Smraza:
We are a leading manufacturer of electronic components for Arduino and Raspberry Pi. We have a professional engineering team dedicated to providing tutorials and support to help you get started.
If you have any technical questions, please feel free to contact our support staff via email.
It was the perfect combination for Raspberry Pi 3B and 2B with good quality and price.(Raspberry Pi Board and SD Card are NOT included)
4 in 1 Starter Accessories Kit for Raspberry Pi 3B and 2B:It includes RPi Clear Case + 5V 2.5A Power Adapter + 3 pcs Heat Sinks(One as standby) + Micro USB Cable with ON/OFF switch.
5V/2.5A Power Supply adapter,Support short circuit protection and overload protection.
Micro USB Cable with ON / OFF switch , No need to pull the cable to restart your PI, just press the button to turn your Pi on and off (Not support data transfer).
3 PCS Aluminium heat sinks(Small one as standby) set are specially designed for Raspberry Pi 3B and 2B to cool down the board,there has stickiness on the back of the heat sinks.


RProtoBuf 0.4.8: Windows support for proto3

By | ai, bigdata, machinelearning | No Comments

(This article was first published on Thinking inside the box , and kindly contributed to R-bloggers)

Issue ticket #20 demonstrated that we had not yet set up Windows for version 3 of Google Protocol Buffers (“Protobuf”) — while the other platforms support it. So I made the change, and there is release 0.4.8.

RProtoBuf provides R bindings for the Google Protocol Buffers (“Protobuf”) data encoding and serialization library used and released by Google, and deployed as a language and operating-system agnostic protocol by numerous projects.

The NEWS file summarises the release as follows:

Changes in RProtoBuf version 0.4.8 (2017-01-17)

  • Windows builds now use the proto3 library as well (PR #21 fixing #20)

CRANberries also provides a diff to the previous release. The RProtoBuf page has an older package vignette, a ‘quick’ overview vignette, a unit test summary vignette, and the pre-print for the JSS paper. Questions, comments etc should go to the GitHub issue tracker off the GitHub repo.

This post by Dirk Eddelbuettel originated on his Thinking inside the box blog. Please report excessive re-aggregation in third-party for-profit settings.

To leave a comment for the author, please follow the link and comment on their blog: Thinking inside the box . offers daily e-mail updates about R news and tutorials on topics such as: Data science, Big Data, R jobs, visualization (ggplot2, Boxplots, maps, animation), programming (RStudio, Sweave, LaTeX, SQL, Eclipse, git, hadoop, Web Scraping) statistics (regression, PCA, time series, trading) and more…

Source link

4 essential skills software architects need to have but often don’t

By | ai, bigdata, machinelearning | No Comments

A look into the unspoken side of software architecture.

Microservices. Continuous delivery. Reactive systems. Cloud native architecture. Many software architects (or even aspiring ones) have at least a passing familiarity with the latest trends and developments shaping their industry. There are plenty of resources available for learning these topics, everything from books and online videos to conference presentations; the ability to go from novice to expert in these areas is right at their fingertips. However, as today’s trends quickly bleed into tomorrow’s standards for development, it’s paramount that software architects become change agents within their organizations by putting into practice the “unspoken” side of their skill set: communication and people skills.

With the arrival of each new year, we all verbalize New Year’s resolutions that fall on deaf ears, including our own. For software architects, these vows may take the shape of getting up to speed on a new form of architecture or finally mastering that cutting-edge technology stack. Either way, it’s the development of these soft skills that define architects as managers—both of large architecture teams and the technology choices they embrace—that’s often neglected in favor of learning the shiniest new technology.

Continue reading 4 essential skills software architects need to have but often don’t.

Source link

An NlogN Parallel Fast Direct Solver for Kernel Matrices

By | ai, bigdata, machinelearning | No Comments
When Matrix Factorization meets Machine Learning:

Kernel matrices appear in machine learning and non-parametric statistics. Given N points in d dimensions and a kernel function that requires O(d) work to evaluate, we present an O(dNlogN)-work algorithm for the approximate factorization of a regularized kernel matrix, a common computational bottleneck in the training phase of a learning task. With this factorization, solving a linear system with a kernel matrix can be done with O(NlogN) work. Our algorithm only requires kernel evaluations and does not require that the kernel matrix admits an efficient global low rank approximation. Instead our factorization only assumes low-rank properties for the off-diagonal blocks under an appropriate row and column ordering. We also present a hybrid method that, when the factorization is prohibitively expensive, combines a partial factorization with iterative methods. As a highlight, we are able to approximately factorize a dense 11M×11M kernel matrix in 2 minutes on 3,072 x86 “Haswell” cores and a 4.5M×4.5M matrix in 1 minute using 4,352 “Knights Landing” cores.

ASKIT is available here:

Join the CompressiveSensing subreddit or the Google+ Community or the Facebook page and post there !
Liked this entry ? subscribe to Nuit Blanche’s feed, there’s more where that came from. You can also subscribe to Nuit Blanche by Email, explore the Big Picture in Compressive Sensing or the Matrix Factorization Jungle and join the conversations on compressive sensing, advanced matrix factorization and calibration issues on Linkedin.

Source link

Breaking a Promise

By | ai, bigdata, machinelearning | No Comments
Yesterday, I talked about eight pricing tiers (you might have 28, who knows, no right/wrong answer here).
  1. Full Price.
  2. Liquidations.
  3. Specials.
  4. Rewards.
  5. Shipping Promotions.
  6. Promotions.
  7. Inconsistent Pricing.
  8. Lies.
It has been my experience that companies that break a promise across pricing tiers are companies that are set up for long-term struggles.

Think about JCP. They moved down the chain, focusing primarily on (6) and (8) for success. Over time, they built a customer base that loved (6).

Then JCP broke the promise. They offered everything at full price (a low price, but still, the low price became the full price). Sales dropped 30%.

Notice that sales didn’t (and haven’t) recover when JCP went back to business as usual.

You could look at Lands’ End under Sears ownership … they promised to never artificially inflate a price to create a sale … Sears undermined that strategy, moving the company from (1) to (3) and (5) and (6). Once you move a customer down the pricing tier hierarchy, it is very difficult to move the customer back up to the top of the hierarchy.

Source link