psankar

I do not have data to back this claim. But the tone of this blog post makes me feel that, the new CEO is an incompetent person who decided to use the vehicle of the covid virus to announce job cuts that he has planned a long time ago.

With the crap ton of marketing buzz in the post, and the casual mention of the layoffs, I don't think the CEO or the new board seem to care all that much about the laidoff employees.

Again, this is my personal feeling and may not be the truth.

https://en.wikipedia.org/wiki/Mukesh_Ambani is the most powerful person in India. More important than the Prime minister or President. He decides who becomes the prime minister, what news people will consume, etc. He owns everything from oil refineries, infrastructure companies, telecom companies, power companies, entertainment companies, news companies etc. He is the wealthiest person in India and one of world's richest.

Since you mentioned about CloudRun, I had one query.

I run docker compose locally for development. For prod, I just use a different docker compose file (with some values changed, for example the postgres database url etc.). I do this from a 5 USD per month droplet/vm. I can launch multiple services like this for my microservices platform. I can use a hosted database solution for another 15 USD per month, to get backups etc. Also I get a generous 1 TB bandwidth and predictable performance for my system as a whole.

In the past I have used appengine and been bitten by their update times (took more than 20 mins for a single code update, things could have improved now). Also I need to write deployment artifacts for each service.

Now is there any benefit that cloud run (or any paas) could offer compared to this ? Would it be not easier to just stay with docker-compose and defer upgrading to kubernetes until you turn profitable or become unmanageable with a single VM ?

All the best

May be it is just me, but there should have been a tldr summary or better title (Arash, co-founder of dropbox leaving) or first paragraph. I was reading it half scared wondering if someone died (due to the recent Corona related news).

Been following your work for a long time now (memcached, livejournal, camlistore/perkeep, etc.) but could not contribute to any of your work sadly (other life priorities, sigh).

Met you in a Golang conf and talked about some client libraries for Go for AWS. You were really helpful and motivating.

Best of luck to whatever you are building. On a personal level, I hope someday I get to work with you in some project :)

https://en.wikipedia.org/wiki/Whataboutism will not take us anywhere.

As a resident Indian, I would say, America is an astrnomically more democratic nation than India. You can make fun of Trump in primetime television, share it in youtube and still live happily. In India, you cannot even talk against a small circle politician or Police and not feel threatened for your life.

The Indian Governments have been notoriously interested in removing any content that may even be slightly critical of them. Look at the Google transparency index for India and the increase in the number of items requested to be taken down in the recent past https://transparencyreport.google.com/government-removals/by...

Also, India is probably the leader in internet shutdowns across the world https://edition.cnn.com/2019/12/14/opinions/india-world-lead...

These new law amendments have got nothing to do with the sovereignty of India but are just tools for censorship and intimidation.

To provide some context, as per the new amendments, the Indian Government wants:

* International companies to setup a physical office in the country, with an executive, who can be legally brought to court, police station etc., if the company has more than 5 million users in India. (Indian population is a Billion strong and has more than 600 million online users)

* Wants automated censorship for unlawful content on their sites

* Wants these companies to remove any content flagged by the Government, within a time period of 24 hours. The companies should also provide user data requested by the Government within 72 hours

> Name may be modern, but Hinduism refers to Sanatan Dharm that is indigenous to India.

Absolutely not. Tamil Sithar songs, Kannada Lingayata principles all oppose caste systems and make fun of vedic practices. Sanatan Dharm is all about discriminating people based on birth/caste.

What absolute non-sense! There is no such religion called Hinduism. An average Hindu Punjabi has more in common with an average Muslim Pakistani (Language, Food, Music, etc.) than a Hindu Tamilian. But the Punjabi/Rajasthani and Tamilian or both clubbed into a single Hindu religion.

A Tamilian sacrifices Goat to his God. But this is not accepted by the Brahminic Hinduism as a way of worship in their temples. But all of these are merged into a false identity of "Hindu".

Who is a hindu according to Indian law ? One who is not Muslim or Christian or Jain. Because there is nothing else that commonly unites them. Hindu is a western term for the land beyond Indus and it is not a religion for all these years (until the politicians decided to use it to suit their purposes). Saivam, Vaishnavism, etc. were the original religions of this land. Not this abomination called Hindutva.

> May I ask why you're running DB and stateful systems in k8s in the first place?

Because it is the easiest way to get our integration tests in CI working fine. We may change in future, if there is a need.

One of our GKE clusters suddenly went missing today (from the console as well as kubectl) and we were scared for some time, panicking how the cluster got deleted.

Google should have put up some kind of alert dialog in the console, saying that some services are experiencing a downtime of some kind.

I have used sqitch http://sqitch.org/ in the past and found it good for postgres. It supports revert and verify too.

Genuine question, what exactly do these maintainers want ? Stallman has already quit his leadership role/position, right ?! Is there something more that he or the GNU project needs to do ?

A little off-topic.

> Our fully preprocessed CodeSearchNet Corpus is available for download on Amazon S3

I am surprised that Github went with S3 for this download. Isn't there a Azure equivalent of S3 for large object storage ? This just shows the dominance of AWS.

Thanks. I will check.

Talking to third party services may not be possible. The deployment scenario would be a private enterprise vpn kind of setup.

GA may not be possible as no thirdparty service could be accessed in the environment where this would be deployed.

Thanks. But it is a bit theoretical. I was looking more on the lines of people's practical experience, with architectural choices, tools that they used, etc.

The analytics are not just limited to web clients. There would be API clients too. The deployment will be in a private enteprise vpn and so talking to external services may not be an option.

I am aware of these tools like cassandra/flink/spark/kafka etc. But I am more curios about the best tools and architectural patterns that work well with each other.

1) The website cannot call to external services, which is the primary reason why we thought about implementing it from scratch.

2) We want to invest in building a good framework to track such metrics systematically over time

3) We have some non-web API clients too. Adblock is not a problem.

4) Accuracy is better. Speed is not that critical and could even be a few minutes delayed.

5) The data will be kept for a few months at least, if not years. The storage costs are not a big problem. This is for enterprise based solution, where the deployment may happen in a private network.

> One last point of architectural difference is this — XFS has B+ tree implementation for searching files and allocating space. This makes searching and fetching the data much faster. OpenZFS has no such feature.

I wish I could downvote this article, just for this one point alone. I wonder what kind of benchmarking the author did to come to the conclusion of "searching and fetching the data much faster" and how much the B+ tree of xfs contributed to it. Speed depends on a lot of factors (copy on write, compression, journaling, crash recovery, etc.) and I wonder how the author narrowed it down to the B+ tree of XFS. Also, there could be variants of various B family trees even within a same filesystem. So I am not really sure, what the author means by "XFS has B+ tree implementation".

If you are a RedHat shop and deploy linux extensively already, then xfs may be a better option, if you are not looking for most zfs features like volume management, etc. Also, installation, support etc. would be much easier.

This article is shallow in many ways technically, but arrives at conclusions strongly. It just seems like a paid article to promote zfs.

May be an article on LWN would be more meaningful and technically competent to make a comparison than this.

I am not a doctor but had some tiredness issues and found that low vitamin d or low thyroid could cause such. Other symptoms include dry skin, depression, inability to focus, muscle pain etc.

Postgresql gets native JSON support (at least since 9.2 onwards) to store schemaless free flowing text. Dynamodb gets transaction guarantees.

There is globalization and intermingling happening on technology too.

On a similar thought, a few years back, C# and Java got `Any` generic types, while Python/JS got static types (via python3 typings, typescript)

This is better than the original article.

Google has terrible support. I used AWS and Google clouds in parallel for a while and the difference was day and night. GCE support issues take anywhere from 2 days to 3 days for the first response (depending on timezone differences), while with AWS someone usually contacts within hours, if I just crib on the twitter account. There are only positive surprises with AWS, while negative surprises only with GCE (like their recent maps pricing debacle, retiring Datastore APIs, etc).

AppEngine was ahead of the competitors a few years back. But it was/is neither as easy as heroku, nor as kept updated as the competitive PaaS solutions. We have beanstalk, lightsail etc. in AWS. Appengine flexenv takes about 10 minutes whenever we change the container sources, and the GCE team believes that it is acceptable :(

Golang was invented at Google but AWS Lambda added support for serverless golang a looooooong time ago, while CloudFunctions from Google is still in a private closed beta which probably none outside Google use.

Also, I have been burnt by Google obsoleting their Datastore APIs and expecting the whole world to move as fast as them, while my apps written for dynamodb continue to run without requiring changes for years.

Also, firebase, stackdriver (opentracing vs observability) etc. seem to go on various directions, with different documentation formats, tutorial styles, dashboard styles, etc. in contrast to AWS services, which offer a more cohesive, uniform experience.

Some of us at $DAYJOB are big fans of google technologies but always end up choosing AWS due to the much superior support of the latter. It often feels to me like Google is in the public cloud business only as an afterthought and not even fully interested in it.

Now you have two "doing" :)

This is exactly the reason why we went away from appengine flex to AWS Beanstalk. As much as a google fan I am, this is unacceptably high and there seem to be no interest by Google to reduce this. Beanstalk isn't that great either but better than appengine flex when it comes to the deployment wait times.

The best keyboard that I have personally used is the old IBM Thinkpad keys. They had a slight curve for accommodating touch typists, instead of the flat keys in the modern keyboards. I could type at 100+ WPM in the old IBM thinkpads even when I am sitting in a dark room. I struggle to reach around 80 wpm in the macbook pro and modern lenovo thinkpads.

The new macbook pro with the touchbar is the worst keyboard that I have ever used and I ended up buying an old model Macbook just because of the keyboard.

I wish Microsoft or Google releases a good developer oriented laptop with proper keyboard (Preferably licensing the old IBM thinkpad keyboard style) and full-size buttons for arrow keys, boatloads of extensible RAM and a gorgeous screen. Apple seem to not care anymore about macbooks :(

I've been using gitlab for a one person company for the last six months or so. The free plan is good for me. There are private repos, CI/CD Integration (pipelines). I am not using Issues or wikis etc. I use pull requests though. I just need git hosting and a decent CI/CD pipeline system.

I don't mind paying gitlab though to avoid the hassle of hosting things myself. But the Bronze plan does not add any value for me, as there is no increase in the CI/CD minutes compared to the free plan. The silver plan is not affordable to me.

Are there any other FOSS tools that HN readers know that provides out of the box support for git hosting as well as CI/CD ? This is just a backup planning in case gitlab decide to reduce or stop the CI/CD minutes.