cpitman

> The flexibility benefit modern Python gives is probably more like 1000x

If true, that would mean that a 1 year C++ project could be done in 2 hours, and a 10 year project in half a week.

We (Red Hat) also released an easy way to do a side by side install of newer tooling without impacting the OS. Python 3 was released on Red Hat Enterprise Linux in 2013, and we currently support Python 3.6.

Here's more on how software collections work and what is released: https://developers.redhat.com/products/softwarecollections/o...

Hah, that happened to me at the Canadian border. When asked what my profession was, I said "Computer Programmer", but my visa was issued for an "Architect". Had fun explaining that I am a "Software" Architect, not a building architect.

I agree, when Kafka is (or can be made) a good fit it simplifies a lot of this. Thanks for the presentation link. I'm seeing companies trying to use Kafka for "Request style" services, ie as just another enterprise messaging system.

Some messaging systems have features to make ordering easier. For example, ActiveMQ has Message Groups (http://activemq.apache.org/message-groups.html) which guarantee a single consumer processes message from the same group and that it receives them in order.

What I'd like to advocate for is a middle ground: using queues at key points in the larger distributed system to reduce the scope of cascading failures and HTTP/P2P messaging elsewhere.

If a request comes in from an external system, and that request is queued, then the external system is already not directly impacted by downtime after the queue. Additional latency in processing after the queue is also hidden by the single queue. After a single queue makes the overall system async the goal should be to increase overall throughput and efficiency, and I don't believe only using queues accomplishes that.

I've seen this play out multiple times now, where the engineers decide they are "doing messaging", and use it for everything, even what could be simple request/response REST services. Tuning the overall system for performance becomes complicated, and the messaging brokers have to be scaled out to handle the multiplied load.

On a less serious note, the logistics game Factorio (https://www.factorio.com/) really drives this point home. In a distributed system/assembly line, really large buffers all over the place makes it harder to detect issues and tune throughput.

Too many people would use the term "refactor" when they really mean "rewrite".

As far as microservices, one benefit is that it makes blurring the lines between domains much harder. A mediocre developer can easily go into a well designed code base and make a mess of it quickly. In a microservice setup, their mess is mostly contained to one domain at a time.

Mostly agree, but partial classes were a far later addition

The article is also about Asmodee purchasing Lookout Games. They bypassed purchasing Mayfair, which was publishing games created by Lookout Game's oversees, and purchased Lookout instead.

I'm surprised they don't purchase it just for the brand. As an American board gamer, I instantly notice games that have the Mayfair logo and expect them to be the type of games I like.

Or a car with a "banker cargo" vs a "barista cargo". When I take a toll road, it is with full knowledge of what my time is worth, and that most people can't justify it.

Yep, this will start yet another bidding war on our area. For the last couple years we've had a constant back and forth of each region trying to claim the new FBI headquarters.

I usually advocate for storing (and therefore versioning) tests with the code under test, but this one complaint may have changed my mind. Backwards compatibility is incredibly important for microservice architectures. The more tests are black box, the more this would work.

Depends what you mean by fast. Anecdotally, I learned to use a computer well before anyone thought to teach me to type. I still type "incorrectly" (basically use only a few fingers, with my hands moving more than they should). So I am definitely slower than someone who is trained to type correctly.

But, I would like to think that I am both good and experienced. I picked up my bad typing habits while learning to program, after all.

100%. I work in consulting, and we spend a lot of time trying to find ways to do this better. But no amount of clearly documented prerequisites, checkpoint meetings to try to make sure they'll be ready before kickoff, or anything else consistently works. In my experience, we often lose 1+ week due to the customer not being ready even though they have acknowledged the prerequisites.

I think it comes down to priorities, and the fire in front of you always has higher priority. That means that onboarding and prereqs are not the priority until the consultants show up on-site and you are paying for them to find other things to do while waiting for the real project to start. Then it is a fire, and everyone starts moving with purpose.

PTC is a lot more than overspeed control, imagine any use case that would be enabled by electronically controlled trains connected to a highly available network.

I'd guess that until PTC becomes standard across the country, new lines may be deployed with out it. The lack of PTC certainly doesn't stop them from running trains on pre-existing lines today.

I've spent almost the last decade involved with designing and implementing the messaging system behind PTC. Every one of these disasters makes me wish it was deployed everywhere. But for better or worse, railroads are incredibly change and risk intolerant. Everything goes through extensive testing before being rolled out into the field, which can add several years to any rollout plan.

Yep, the US Space and Rocket Center is a museum right outside the base. The Davidson Center for Space Exploration (seen here https://www.rocketcenter.com/admission) has a Saturn V rocket broken down into stages. I had an hour to run through there while on a work trip, wish I could have spent more time.

This is assuming the breach is stealing a disk though, right? If the encrypted storage is part of an application, and I infiltrate the application or can trick it into serving me data, then the encryption cannot help. Any runtime breach isn't really prevented by encryption at rest.

No, that wasn't me. I think you are incorrect, though. Fox was only an example, this story was carried by most national outlets. And I've never met anyone who changes their mind because the majority is against them and is silencing them. It just brews resentments and causes people to dig in.

If people changed their minds just because everyone else disagrees with them loudly and violently, how would any previous social change have come about? New thoughts and perspectives always start as minority views. So majority consensus must not be enough to change someone's mind.

Also, this really isn't about winning. We're all losing right now.

It is not just those people being silenced though. It is people who stayed home, maybe their views are not quite as radical as someone who would protest, or maybe they tend to be more politically apathetic (ie most people).

When those people watch Fox News (http://www.foxnews.com/tech/2017/08/14/twitter-users-identif...) and see what is happening to those people for their speech and actions, they are being silenced.

These are also probably the people more amenable to changing their views. It's got be pretty hard to swing someone with a tiki torch marching down the street, but someone who feels sympathetic to their views could maybe be reasoned with if they would talk about it.

The White Nationalist march on Charlotte is actually an example of voices being silenced. Photos of people there were distributed, their names were found (sometimes incorrectly), and their places of work contacted. Then they were fired.

If people see that people who share the same views are losing their livelihoods, they are more likely to be silent. Only the independently wealthy wouldn't consider losing their job risky.

And just to be clear, I in no way support their views, desires, or actions. But claiming the goal isn't to silence them is disingenuous.

Yes! If everyone knew it was a popular election, the results would be drastically different. Candidates would actually campaign and get out the vote in states they never even think about now, and voters in states that are not swing states would be more motivated to vote.

Saying Clinton won the popular vote is just wishful thinking.

Oh geez, thanks both for pointing that out.

I don't see that. It has tools for running the benchmarks yourself, but that doesn't give you permission to publish benchmark results. And even if it did, a blog post from an employee cannot give you permission to do something that the EULA takes away.

This has often been abused by these companies to attack open source software that allows benchmarking (because what kind of open source license would ban that?). The company can perform a "benchmark" of their product and the open source product. They'll then release the results as a whitepaper. Even if they perform the benchmark correctly, they can always choose which results to actually publish.

But since the proprietary software has a "no benchmarking" clause, open source projects cannot respond to the whitepaper by performing their own benchmarking. They would need the permission of the proprietary vendor!

For example, here is an IBM blog post comparing the performance of IBM MQ and Apache ActiveMQ (https://webspherecompetition.wordpress.com/2015/03/12/ibm-mq...). I've tried to find a copy of the IBM MQ EULA to link, but cannot find one anywhere. But last time I reviewed it (several years ago) I believe it also had a "no benchmark" clause.

Isn't this a common objection by teacher unions when pushing against using metrics for retention and compensation?

These policies go two ways. Companies need to be able to delete data eventually, but they also need to make sure that employees are not deleting data on their own in order to hide illegal activity. Having a retention policy of X months then cuts both ways. If you delete data before 18 months, well that looks awfully suspicious.

If I vote for a state government that maintains higher state taxes, why should that decrease what I owe to the nation? Am I getting less benefit from the federal government? Also, those states can always lower their taxes in the future.

That seems to agree with me? Mainframes are still in use everywhere. However, that statistic doesn't imply that those customers build their brand new greenfield capabilities on those mainframes.