Original image

World War I Centennial: Bulgarian Offensive Halted at Chataldzha

Original image

The First World War was an unprecedented catastrophe that killed millions and set the continent of Europe on the path to further calamity two decades later. But it didn’t come out of nowhere.

With the centennial of the outbreak of hostilities coming up in 2014, Erik Sass will be looking back at the lead-up to the war, when seemingly minor moments of friction accumulated until the situation was ready to explode. He'll be covering those events 100 years after they occurred. This is the 44th installment in the series. (See all entries here.)

November 16-18, 1912*: Bulgarian Offensive Halted at Chataldzha

Bulgarian bayonet charge at Chataldzha

After scoring stunning victories over the Turks at Kirk Kilisse, October 22-24, 1912, and Lule Burgas, October 28-November 3, the Bulgarians pursued the retreating Turks until the latter made their stand about 20 miles west of Constantinople at Chataldzha (Turkish, Çatalca) – where a line of hills cuts the peninsula north-south and the land is made even narrower by two coastal lakes, offering an excellent defensive position.

By mid-November, three Turkish armies (I, II, and III) were dug into a network of trenches on the hills, with three reserve armies (1, 2, 3) forming a second line of defense behind them; altogether around 138,000 Turkish troops faced approximately 173,000 enemy troops in the 1st and 3rd Bulgarian armies. The Turkish commanders frantically scraped together artillery from Constantinople and its surroundings, and also drew firepower from a Turkish battleship and cruiser in the Sea of Marmara, while the Bulgarians struggled to bring their own artillery up over muddy Balkan roads.

Chataldzha was going to be a last stand for the Turks: if Constantinople fell to the Bulgarians, it would only be a matter of time before Europe’s Great Powers descended to carve up the rest of the Ottoman Empire. Many Turkish soldiers at Chataldzha were now fighting to protect their own homes and families, and while morale was probably low, they were more determined than ever. The approaching Bulgarians, by contrast, were exhausted by hundreds of miles of marching and hard fighting in two major battles, and their ranks were further thinned by an epidemic of cholera, which erupted in the Turkish armies and quickly spread to the Bulgarians.

While the odds didn’t quite favor the Bulgarians, there was never any doubt that they would try to conquer Constantinople regardless. Slavic culture, shaped by its long association with Byzantium, venerated Constantinople as the capital of the Byzantine Empire as well as the seat of the Orthodox Christian patriarchate, and it was simply too big a prize for the Bulgarians to pass up. Among other things, capturing the imperial city would legitimize King Ferdinand’s pretentious claim to the title of “Tsar” (Ferdinand had a full-length portrait of himself dressed as a Byzantine emperor, and kept the regalia in his closet just in case he needed to make a triumphal entry to Constantinople some day). On Saturday, November 16, 1912, Tsar Ferdinand ordered the attack, and his second-in-command, Mihail Savov, assured reporters: “Gentlemen, we shall be in Constantinople in eight days.”

Over-confident after their earlier triumphs, the Bulgarians decided to use the same simple plan of attack that delivered their initial victories – a heavy artillery barrage followed by a mass frontal assault on the Turkish positions by Bulgarian infantry. Beginning at 5 a.m. on Sunday, November 17, Bulgarian artillery shelled Turkish positions while Bulgarian troops advanced towards the Turkish trenches in a heavy fog. However, after progressing several hundred yards the Bulgarian infantry came under punishing Turkish artillery fire and naval bombardment. With Bulgarian artillery mostly failing to inflict significant damage on the enemy, several human wave-style attacks on the Turkish trenches were repulsed, with major Bulgarian casualties from massed Turkish rifle fire and machine guns; one military observer from England described the Bulgarian attacks as “the most futile and wasteful thing, he had ever seen in his life.” During the course of the second night, November 17-18, a Bulgarian assault actually managed to occupy a section of the Turkish forward trenches, but was quickly driven out by a Turkish counterattack before the Bulgarian forward units could receive reinforcements, foreshadowing seesawing trench warfare in the Great War to come.

Nonetheless the fighting continued amid heavy rain and fog – the First Balkan War seemed to be one long downpour – through the night of November 17 and into the night of Monday, November 18, when Tsar Ferdinand and Savov finally called off the assault. Total casualties came to around 12,000 killed, missing, and wounded for the Bulgarians, versus around 5,000 killed, wounded, and missing on the Turkish side (both armies also suffered heavily from cholera, which killed around 600 Bulgarian soldiers and 1,000 Turkish soldiers, and incapacitated many more, around the time of the battle).

The European Great Powers were all relieved by the Bulgarian defeat at Chataldzha, which meant the Ottoman Empire would survive at least in the near term – thus putting off the day when they would all have to scramble to claim their own portions of the dismembered empire. Even the Russians, while supposedly supporting the Bulgarians, were actually secretly relieved that they failed to capture Constantinople, which the Russians wanted for themselves.

Serbian Forces Reach the Sea

Meanwhile a new diplomatic crisis was brewing between Russia and Austria-Hungary over the issue of Serbian access to the sea in the western theatre of war. This was no longer just a hypothetical possibility: on November 17, 1912, Serbian forces reached the Adriatic Sea at Alessio (Albanian: Lezhë), about 50 miles north of Durazzo, meaning the crisis was about to begin in earnest.

Austria-Hungary’s foreign minister, Count Berchtold, was determined to prevent Serbia from gaining access to the Adriatic Sea, which would make Serbia less dependent on Austria-Hungary and threaten Austria-Hungary’s access to the Mediterranean; with Serbian troops standing on the shores of the Adriatic, Berchtold had to act now or risk defeat with each passing day. Meanwhile the Russians had to decide how far they were willing to go to back their Serbian clients against Austria-Hungary in the dispute over a port on the Adriatic Sea. Austria-Hungary and Russia were preparing to face off, raising the possibility of a much larger war.

See previous installment, next installment, or all entries. This installment should have run yesterday.

Original image
iStock // Ekaterina Minaeva
Man Buys Two Metric Tons of LEGO Bricks; Sorts Them Via Machine Learning
May 21, 2017
Original image
iStock // Ekaterina Minaeva

Jacques Mattheij made a small, but awesome, mistake. He went on eBay one evening and bid on a bunch of bulk LEGO brick auctions, then went to sleep. Upon waking, he discovered that he was the high bidder on many, and was now the proud owner of two tons of LEGO bricks. (This is about 4400 pounds.) He wrote, "[L]esson 1: if you win almost all bids you are bidding too high."

Mattheij had noticed that bulk, unsorted bricks sell for something like €10/kilogram, whereas sets are roughly €40/kg and rare parts go for up to €100/kg. Much of the value of the bricks is in their sorting. If he could reduce the entropy of these bins of unsorted bricks, he could make a tidy profit. While many people do this work by hand, the problem is enormous—just the kind of challenge for a computer. Mattheij writes:

There are 38000+ shapes and there are 100+ possible shades of color (you can roughly tell how old someone is by asking them what lego colors they remember from their youth).

In the following months, Mattheij built a proof-of-concept sorting system using, of course, LEGO. He broke the problem down into a series of sub-problems (including "feeding LEGO reliably from a hopper is surprisingly hard," one of those facts of nature that will stymie even the best system design). After tinkering with the prototype at length, he expanded the system to a surprisingly complex system of conveyer belts (powered by a home treadmill), various pieces of cabinetry, and "copious quantities of crazy glue."

Here's a video showing the current system running at low speed:

The key part of the system was running the bricks past a camera paired with a computer running a neural net-based image classifier. That allows the computer (when sufficiently trained on brick images) to recognize bricks and thus categorize them by color, shape, or other parameters. Remember that as bricks pass by, they can be in any orientation, can be dirty, can even be stuck to other pieces. So having a flexible software system is key to recognizing—in a fraction of a second—what a given brick is, in order to sort it out. When a match is found, a jet of compressed air pops the piece off the conveyer belt and into a waiting bin.

After much experimentation, Mattheij rewrote the software (several times in fact) to accomplish a variety of basic tasks. At its core, the system takes images from a webcam and feeds them to a neural network to do the classification. Of course, the neural net needs to be "trained" by showing it lots of images, and telling it what those images represent. Mattheij's breakthrough was allowing the machine to effectively train itself, with guidance: Running pieces through allows the system to take its own photos, make a guess, and build on that guess. As long as Mattheij corrects the incorrect guesses, he ends up with a decent (and self-reinforcing) corpus of training data. As the machine continues running, it can rack up more training, allowing it to recognize a broad variety of pieces on the fly.

Here's another video, focusing on how the pieces move on conveyer belts (running at slow speed so puny humans can follow). You can also see the air jets in action:

In an email interview, Mattheij told Mental Floss that the system currently sorts LEGO bricks into more than 50 categories. It can also be run in a color-sorting mode to bin the parts across 12 color groups. (Thus at present you'd likely do a two-pass sort on the bricks: once for shape, then a separate pass for color.) He continues to refine the system, with a focus on making its recognition abilities faster. At some point down the line, he plans to make the software portion open source. You're on your own as far as building conveyer belts, bins, and so forth.

Check out Mattheij's writeup in two parts for more information. It starts with an overview of the story, followed up with a deep dive on the software. He's also tweeting about the project (among other things). And if you look around a bit, you'll find bulk LEGO brick auctions online—it's definitely a thing!

Original image
Name the Author Based on the Character
May 23, 2017
Original image