NIFTY & US 10 Year Yield have a high degree of positive directional correlation. Anyhow, they are diverging at the moment though the divergence can go on for a long period of time as well, last time around this divergence did not work well for NIFTY & it got pulled down.

Bhu Varaha Swamy Temple

Bhu Varaha Swamy temple is a Hindu temple, located at Srimushnam, in the South Indian state of Tamil Nadu. Constructed in the Dravidian style of architecture...

the temple is dedicated to Varaha (Bhu Varaha Swamy), the boar-avatar of the god Vishnu and his consort Lakshmi as Ambujavalli Thayar.

The temple had contributions from Medieval Cholas of the 10th century with later expansions by Thanjavur Nayak king Achuthappa Nayak.

A granite wall surrounds the temple, enclosing all the shrines and the temple tanks. There is a seven-tieredrajagopuram, the temple’s gateway tower.

Six daily rituals and three yearly festivals are held at the temple, of which the Chariot festival....

celebrated during the Tamil month of Vaikasi (April–May), being the most prominent. The festival also symbolises Hindu-Muslim unity in the region – the flag of the chariot is provided by Muslims; they take offerings from the temple and present to Allah in the mosques.

Source & Pic credit : Anusri

A thread on HN about bad code in legacy projects both makes me think how little we've learned as a discipline over the years and, honestly, how little credit we give ourselves for some pretty major

Fun going down this list and thinking: "Hmm, plausible at a well-run modern software shop", "Hmm, possible, but requires implausible tradeoffs", "Literally disallowed by languages", and "If you were to attempt doing that our test suite wouldn't let you merge."

I think we as an industry celebrate (not quite the right word) failure too much and don't celebrate success nearly enough. There is no DailyWTF for competent execution, word of which generally stays pretty local to the source while incompetence passes into legend.

Alrighty let me try to thread the needle on being the change I want to see in the world while not giving away anything that will get me in trouble:

Ruby has wonderful developer ergonomics. Typed languages are easier for machines to guarantee the correctness of. We built a type checker for Ruby (and I believe it is slated for OSS release sometime).

c.f.

Thread: History of Mogadishu and it’s conquest by Imaam Al Yaqubi

One of the First Imaams of the Hiraab Imamate was Mohamed Ibn Imaam Ahmed ibn Imam Ahmed Ibn Imaam Mohamoud Ibn Imaam Cumar Halol Al Yacquubi (Abgaal) during the year 1097 hijri.

After the collapse of the Ajuuran Abgaal nomads from the interior had started moving to take over the strategic port town of Mogadishu which was then under the muzzafar dynasty. By the 1600s Abgaal warriors had captured the city and set up their base in shingaani.

Mogadishu at the time was a large and well known port town. The Abgaal clan taking over control allowed them to obtain political power and world wide recognition.

Under Imaam Yaqubi’s time in power it’s reported that French merchants had arrived at the port in 1701
and stayed for 11 days. They had attempted to take the city but were successfully repulsed.

Important paper from Google on large batch optimization. They do impressively careful experiments measuring # iterations needed to achieve target validation error at various batch sizes. The main "surprise" is the lack of surprises. [thread]

https://t.co/7QIx5CFdfJ

The paper is a good example of lots of elements of good experimental design. They validate their metric by showing lots of variants give consistent results. They tune hyperparamters separately for each condition, check that optimum isn't at the endpoints, and measure sensitivity.

They have separate experiments where the hold fixed # iterations and # epochs, which (as they explain) measure very different things. They avoid confounds, such as batch norm's artificial dependence between batch size and regularization strength.

When the experiments are done carefully enough, the results are remarkably consistent between different datasets and architectures. Qualitatively, MNIST behaves just like ImageNet.

Importantly, they don't find any evidence for a "sharp/flat optima" effect whereby better optimization leads to worse final results. They have a good discussion of experimental artifacts/confounds in past papers where such effects were reported.