Recent films that I can remember: John Wick 4, Top Gun: Maverick, NOPE.
All of them had the forced break.
1,028 karma · joined January 24, 2022
Recent films that I can remember: John Wick 4, Top Gun: Maverick, NOPE.
All of them had the forced break.
> That practice, according to the agency, is in large part responsible for inflating demand for the visas to a record high this year, with 781,000 entries into the lottery for 85,000 visa slots.
> About 350,000 applicants for H-1B visas submitted one entry into the lottery this year, compared with about 307,000 last year.
> Some 96,000 people submitted multiple visa entries, for a total of about 408,000 entries.
> About 478,000 people competed for visas last year, a record at the time.
So, 350,000 (single entry submission applicants) + 96,000 (multiple entry submission applicants) = 446,000 unique applicants compared to 478,000 unique applicants last year.
That's 6.7% fewer unique applicants.
ONDC's charter is quite ambitious and it is trying to become the network for ecommerce in the country.
While it is absolutely not a contender to the established ecommerce players in terms of reach, spread, depth etc., yet, it could be a gamechanger if it gets adopted far and wide.
I don't for a moment claim this is perhaps the cause for Amazon to scale down ops in India. Just pointing a broader trend and strategic threat.
[1]: https://ondc.org/
I was in the market for a 43" TV. Needed a Samsung. I checked two reputed local electronics stores (from where I have made purchases earlier; including the store where I bought my current Samsung 32" TV I was replacing after 13+ years).
Both stores quoted a price over the phone. The store I bought my 32" TV came in INR 700 (around $9 USD) lower. This was a 43" Full HD TV.
I then checked amazon.in website. They offered a Samsung 43" UHD TV (i.e., better resolution) at a price that was INR 100 lower than the best price I got from the physical stores. (It was a Limited Time deal, but one that met my requirement).
The product was Fulfilled by Amazon and Prime delivery eligible. I got the TV delivered this morning and Samsung is sending their tech for installation tomorrow. (Amazon handles this seamlessly -- they do for a lot of partner brands).
I would say they met my expectation (on delivery). I'll update after installation by Samsung.
My summary of Amazon (India) experience:
- They are great if you know exactly what you want
- The Prime eligible products Fulfilled by Amazon are good
- I have even returned some products before and so far, that experience has been exemplary. I have no complaints.
I generally avoid other marketplace sellers on Amazon.
(Personal experience)
I joined a startup (privately held) in a senior position. Well comped in cash and bonus. No stock or options. Years later, the company was acquired by a larger company. The owner/CEO who sold the company granted stocks to the senior leadership and I got stock. This was exchanged for stock in the acquiring company.
The acquirer was a funded by strategic investors and venture funds and not IPOd yet. They went public 2 years later. I started getting stock option grants each year.
At that company, the stock option grants were reserved for performance. Let us say people were rated on performance as A (Exceeds), B (Meets), C (Meets some), D (Needs improvement). A would get x% over and above of eligible bonus. B will get eligible bonus, C will get eligible bonus less y% and D doesn't get the bonus. There's bell curves and all that jazz applied for each level (actually sets of levels that are functionally similar eg: Director/Sr. Director would be looked at together and similary VPs/SVPs together, EVPs/Presidents together etc.,).
The A's were eligible for stock option grants depending on tenure (i.e., after xx years of service in the company). The quantum of option grants depended on whether the A's were also eligible for promotion. For eg: a director (rated A) getting promoted as VP would be eligible for an additional option grant quantum as part of the promotion whereas a director (rated A), but not eligible for promotion would get lesser than the first one.
The board has a compensation committee that lays out these rules for "Key Managerial Personnel" eligible for stock/option grants.
Edit: The grants vest over 4 years. Intended to keep the key personnel for that period at least and their behaviors aligned to the shareholders.
(Anecdotal)
I have also heard of companies that grant stock options to a broad base of employees with the eligibility being tenure and performance.
I have always heard it to be a differentiated grant.
An LLM is useless (or not as useful) in OCR for forms where we are trying to extract the name "John Smith" from a name field whereas a HMM trained specifically on name fields may be able to do a better job.
LLMs may perform better as a post processing step for running text (such as book pages).
Is the article about previous qualitative callouts they used to make that were India focused activites that are now deemphasized?
[1]: https://s2.q4cdn.com/299287126/files/doc_financials/2023/q1/...
[2]: https://s2.q4cdn.com/299287126/files/doc_financials/2022/q4/...
Most Indian regional language and Bollywood Hindi films (even the modern ones) run around 2h30m to 2h40m. The older films were even longer.
RRR, the Indian film that captured a lot of viewership around the globe had a runtime of 3h2m.
Most screenwriters and filmmakers in India design around the Interval-block to place a major plot-point just before the interval card is shown.
It is funny that recent Hollywood movies don't come with that. And even when a film is 2h+ long. Indian movie theatres just forcibly introduce a break in Hollywood films so that they can send people to the concession stands and restrooms. Sometimes, this happens mid-dialogue it is funny (and irritating at times).
EU statute has both monthly reporting, yearly and 3 year trends for determining specific applicable rules for large operators.
Example:
The regulation that Apple is trying to comply with EU Digital Services Act [1], [2]
The EU Digital Markets Act [2], [3]
(India) Information Technology (Intermediary Guidelines and Digital Media Ethics Code) Rules, 2021 (updated 06.04.2023) [4]
(Germany) Network Enforcement Act [5]
(France) Law requiring large social media operators to remove hate speech, crimes against humanity, sexual harassment etc., [6]
[1]: https://digital-strategy.ec.europa.eu/en/policies/digital-se...
[2]: https://commission.europa.eu/strategy-and-policy/priorities-...
[3]: https://ec.europa.eu/commission/presscorner/detail/en/ip_22_...
[4]: https://www.meity.gov.in/writereaddata/files/Information%20T...
[5]: https://germanlawarchive.iuscomp.org/?p=1245
[6]: https://www.assemblee-nationale.fr/dyn/actualites-accueil-hu...
The DPLA provides the devs with access to Apple IP.
The courts have ruled that Apple is entitled for some compensation under the IP compensation legal theory. The Appeals court left what some means as ambiguous.
Until some jurisdiction finds the Apple app store as a monopoly and forces Apple to open up ios app distribution to 3rd party app stores, the Netherlands situation is likely to continue where Apple takes the stance that the value of ios distribution for apps is 27% to 30%; 27% for alt payment methods and 30% for using Apple in-app payments. Apple's stance is ios distribution is worth at least 27% of app revenues.
As a very specific case, an app developer targeting the India market could now use the UPI payment method (which is free for up to approx. $1500 per transaction) and pay only 27% or use an alternate payment provider that charges 1.5% transaction fee for a total cost of 28.5% instead of 30% and being forced to use Apple in-app-payments. [This is assuming Apple amends their DPLA with the 27% similar to Netherlands for alternate payment methods around the world].
They probably are worried about the cannibalization of the subscription base and hence caught in the Innovator's Dilemma.
In any case, it is good you didn't have to go through a DC recovery during one of the worst disasters in the 21st century.
The question I keep asking in all DR planning sessions/table top exercises is - what would we do if we had a situation like what happened in Fukushima or in Chennai 2015. In both cases, flooding caused failure of backup power generators. Also, what do we do when we have all or partial resources, but are faced with a denial-of-premises situation (what I faced).
Our DC was intact, but the building and access was cut-off. We lost the backup diesel power generators in the flooding. Of course, grid power was cut-off.
Our DC operating team managed to shutdown all the servers and racks cleanly before UPS power was completely drained. The 4 engineers and 2 security guards then swam out of the compound in chest high waters. (I am not kidding).
When the rains subsided and the flood waters receded after a couple of days, we had to plan the restart. The facility still had to be certified by health and safety, but we needed to get the datacenter back up.
A secondary operations site that would remote-connect to the DC was brought up in 1 week since we estimated the rains to potentially continue for a few more days and cause interruptions. But the critical item for the plan to work was getting a new backup power setup. We rolled in a truck-mounted diesel generator and positioned it in the highest point in the campus (also close to our building tower that had the DC) and ran power cables to it (we had to source this and it was a challenge to do it with the time crunch and the rains).
We moved staff to other cities by bus (airport was shutdown) as part of our recovery plan, but we still needed connectivity to our DC for some of the critical processes.
Long story short, it worked.
I'll never forget the experience and the scars from this war story.
$NFLX brought in customers with the Sharing is Caring campaign. If at that time they patted themselves on the back for customer growth, they should hold themselves responsible for the about face and customer exodus when they have changed their stance to Sharing is a crime.
> “Certain costs associated with corporate initiatives supporting consumer-facing activities, previously reflected in unallocated corporate costs, are now allocated to Google Services; and centrally-managed shared research and development activities, including our shared developer tools, are now allocated based on an updated measure of the relative benefit derived from the services,” Alphabet said in a filing.
> “As a result of these changes, more of the previously unallocated corporate costs are allocated to our segments, and more of certain previously allocated costs are allocated to our consumer-facing Google Services products and less to Google Cloud enterprise products.”
I guess the new tax rule on accounting of software developer salaries as amortized cost over 5 yrs is expected to positively impact profitability. Looks like they used this to also reallocate some central costs to various business units.
From the linked article [1]:
> December 2018: Google+ Bug Exposes 52.5 Million Users’ Data Google+ faced its second big breach of 2018 when a November update created an API bug that exposed data from 52.5 million Google+ accounts. Google fixed the bug within six days, and moved up Google+’s burial date from August to April 2019.
> Google originally decided to terminate Google+ after another breach became public earlier in 2018
and an earlier Google+ bug that was reported in WSJ [2]
> Google Exposed User Data, Feared Repercussions of Disclosing to Public
> Google opted not to disclose to users its discovery of a bug that gave outside developers access to private data. It found no evidence of misuse.
[1]: https://firewalltimes.com/google-data-breach-timeline
[2]: https://www.wsj.com/articles/google-exposed-user-data-feared... (Oct, 2018)
So, if the Netherlands dating apps case ruling is any indication (Apple discounted the 30% to 27% for dating apps in Netherlands for using alt payment methods in that suit), Apple is likely to do something similar here in cases where their In-app-payments are not used.
That compound failure mode (H + O_2 explosive mix upon structural failure/leak) doesn't exist with Helium. So, I respectfully disagree that the fears are outdated.
A survey of NASA's experience [1] is informative. From the linked article:
> Even small amounts of liquid hydrogen can be explosive when combined with air, and only a small amount of energy is required to ignite it. Both its explosiveness and the extremely low temperatures involved make handling it safely a challenge. It has taken a lot of experience, including the experience of dramatic failures, to teach us just how dangerous it is and how carefully we need to treat it.
[1]: https://www.nasa.gov/pdf/513855main_ASK_41s_explosive.pdf
While the contours of how sideloading is going to unravel is unknown at this time, it is possible that Apple may still use some form of app notarization (perhaps an API for 3rd party stores and require that the stores use those APIs -- similar to their stance on webkit).
Then, similar to the Netherlands case where dating apps were allowed alternatives to InAppPayments, but charged 27% commission, Apple might charge a discounted commission (relative to the 30%) for sideloaded apps.
They will look to get their cut.
I don't like it, but that seems to be where this is headed.
The district court and the appeals court both agree that Apple is well within its rights to charge developers $99 initially and x% of revenues, under the principle of what they are calling IP compensation.
The district court did not completely agree to Apple's argument that 30% is fair (they expressly rejected it) and the appeals court observed that Epic's tactic on this argument was legally flawed and mentioned something to the effect of Epic should have used a different argument (and since they didn't, we are not ruling on this). To me, this looks like a vulnerability for Apple that they could be attacked by a different plaintiff using a different legal theory.
> We now turn to Apple’s cross-appeal, beginning with its arguments concerning the UCL. The district court found that Epic suffered an injury sufficient to confer Article III standing, concluded that Apple’s anti-steering provision violates the UCL’s unfair prong, and entered an injunction prohibiting Apple from enforcing the anti-steering provision against any developer. Apple challenges each aspect on appeal. We affirm.
i.e., the District Court found Apple to be violating CA Unfair Competition Law which Apple had appealed and the US Court of Appeals for the Ninth Circuit concurs with the district court.
Consequence of this ruling is limiting the ability of app developers to communicate the availability of alternative payment options to iOS device users is unlawful. As a result, alternate payment methods can no longer be restricted by Apple.
The question of commission (30%) still remains because the ruling is silent on it except for observing what the district court said.
> Because the district court accepted only a general version of Apple’s IP-compensation rationale (that Apple was entitled to “some compensation”), Epic at step three needed only to fashion a less-restrictive alternative calibrated to achieving that general goal, instead of one achieving the level of compensation that Apple currently achieves through its 30% commission.
My reading of this is that the 30% commission can come under challenge from another plaintiff. Epic took a different tactic on this specific argument and failed to establish it. However, the Appeals court's observaton seems to indicate there is legal room for a challenger to attack this.
[1]: https://cdn.ca9.uscourts.gov/datastore/opinions/2023/04/24/2...
But, it is done at the time of writing the PR or requirement. For eg: "Add paid subscriptions" feature request. If you have, say, a markdown document that outline the feature request, it would be a simple matter to have a script pull specific tags to build your release note. A ProdMgr would own the release note anyway. So, it doesn't really matter.
My 2 cents, anyway. Others might have a different view.
> Earlier this year, Tesla made a first effort to try to get those vehicles off the perk by offering owners an extra $5,000 discount if they trade in a Model S and Model X with unlimited free Supercharging.
> Now Tesla is doubling down on this effort by launching what it calls a new “Ownership Loyalty Benefit.”
> The automaker is offering those owners six years of unlimited Supercharging when buying a new Model S or Model X by the end of the quarter.
In the article How much does it cost to charge a Tesla? [1] the Cost to charge a Tesla Supercharger/DCFC is in the range of $5.28 per 100 miles (Model Y) to $5.97 per 100 miles (Model 3) to $6.37 per 100 miles (Model X) to $5.88 per 100 miles (Model S Plaid).
Lets take a blended average of $5.80 per 100 miles.
Average miles driven per year (CA) is 14,435 miles. [2]
That gives a supercharger charging cost of $837.23 per year. For 6 years, that is a cost of $5023.38 without considering time value of money. If we use the national average from [3], Tesla Model Ys are driven 13,500 miles and at the average Model Y charging cost of $5.28/100 miles that is $712.80 or $4276.80 over 5 years.
And customers did not bite the $5k offer in expected numbers to surrender the free supercharger perk. Perhaps, Tesla should look at upping the offer.
[1]: https://electrek.co/2021/10/27/how-much-does-it-cost-to-char...
[2]: https://www.caranddriver.com/auto-loans/a32880477/average-mi...
[3]: https://insideevs.com/news/587352/tesla-modely-average-annua...
> When a bad Pub/Sub message was pulled into the binary, an unhandled panic would crash the entire app, meaning web, workers and crons all died.
Several years ago while working on a Message Queue based application, we had a poison message, one where the consumer process would read the message and then die because the message was malformed and then the message would get back to the queue and due to the FIFO nature of the queue, the retries will also fail.
Ideally, there should be enough checks to ensure that a poison message can never be created. But if edge cases cause this issue, the handler should have some way of knowing that it is retrying the message for the n-th time and then if it fails, move it to a dead-letter-queue and report failure on handling that message.
Unfortunately, this pattern of issues are quite common. So, we fully expect this and design around it.
Then, any message in a dead-letter-queue gives you an example of a crash-inducing message which can then be debugged for root-cause.
VSCode is a good example. After the latest updates, the Recent Changes are surfaced when we open the editor.
That combined with a What's New? link should answer this question any time.
If the issue is customer/end-user view versus internal view of changes, you could put them in a separate section of the Recent Changes release note.
For status tracking in the pre-release/dev/testing stages (for eg., to answer your when is featureX shipping? question, wouldn't something like GitHub tags suffice?
WARNING
This PNG image <link> is an example of a compressed image file that might expand to a huge size upon decompression. PLEASE DO NOT DOWNLOAD IT. We are not responsible for consequences if you do not follow this instruction.
and someone does that anyways, they are to blame, aren't they?
Especially downloaders who claim they are entitled to download it and hosters must use their proprietary opt-out scheme (like the ones on the original linked article) should also learn that they should follow the hoster's instruction and NOT download the zip bomb.
(And the hoster also lists that specific file in robots.txt)