stargrazer

Let us clarify/summarize: ssh didn't go away, it simply became less visible, and remains the secure channel component of gaining access to remote screen/tmux sessions.

Are you simply closing the airport? Is a replacement planned? Or does that traffic simply get redirected to the San Francisco airport?

The article closes with 'think big', but I think is incomplete in scope.

The high end office chair companies will usually offer a try before buy deal.

The Aeron chairs by Herman Miller are popular in some re-insurance companies. I have used the Mirra 2 chair, but prefer the Aeron.

The Aeron, unless purchased used, probably won't fit in your budget, but breaking the budget on such an important item as this will probably be of value.

I came across nordvpn from several angles, and, correct me if I'm wrong, is not chinese owned or influenced. It is what I use on a linux box, using a manual config (their auto install doesn't work quite so easily if the correct packages are not already present (openvpn, etc)). I also manually create the manual null default route. I should write a blog article about the nuances of my manual config.

If prices maintained their current states of affairs, however, a problem is: supply and demand. Lots of supply, demand goes down, price goes down. Subject to a few other variables in the equation regarding cost of mining, distribution, manufacturing subsidiary products, ...

What does it do? Is there a documentation link?

I've installed Linux on Sager/Clevo. Unfortunately, can't seem to buy them without Windows pre-installed.

There is System 7 with Linux native laptops.

So... other than using std::function and other than something in Qt, what would be a more efficient C++[11,17] native way of doing a callback?

Coming from Linux and Windows: the Mac touch pad works well. Once you learn the strokes and finger combinations, no mouse is needed. It becomes natural in use. And the control panel has various knobs if you don't like some of the defaults.

As an addendum, should the article not have the link, a companion article on the same site:

https://developers.redhat.com/blog/2018/10/22/introduction-t...

Here is an article [0] which may define the background to some serious concerns about Huawei. It was almost like Microsoft's purported 'embrace, extend, extinquish' philosophy.

"Many Canadians believe that Huawei stole the core technologies and business strategies from Nortel and used that knowledge to drive Nortel out of world markets and into bankruptcy. "

[0] https://www.assemblymag.com/blogs/14-assembly-blog/post/9063...

Define 'simpler'?! A monolith is easier to build.

When you get into microservices, you get into additional layers of complexity.

The primary one being that you now need to get into inter-service communications, which means serializing/de-serializing messages across the wire, and building in structure to handle the associated failure modes.

Short answer: "don't be silly, too many hard things in the Linux Kernel get correct in Rust"

What they should look into is trying to build eBPF and XDP in Rust. Then you get that compiled language speed and safety in the kernel.

A few quotes:

"I view OpenBMC as on its knees with tears streaming down its face begging for someone to please rewrite me in Rust, because if we're going to have a BMC that's going to hang out a socket over the internet, God forbid, or even over the network, I want that thing to be in Rust."

"That the beauty of Rust is its ability to interoperate, cooperate, with native systems or C-based systems; allow for vistas to open up."

Must be somewhat reasonable, FreeCAD uses the library.

Use https://github.com/digitalocean/netbox to record your network information. Then you can create topology maps with dot/graphviz. https://netbox.readthedocs.io/en/latest/additional-features/...

Someone has a tutorial on using graphviz directly: https://mikegriffin.ie/blog/20110308-a-graphviz-tutorial

For 'dot', I was considering more the machine inter-relationshps. Given ip addresses and subnets and interfaces in the configuration files, these relationships can be diagrammed given the 'sources of truth'.

SaltStack files have state files (commands and general configs) controlled by a 'top' file which lists hosts and their configurations. SaltStack files have pillar files (configuration parameters and values unique to device being configured) controlled by a 'top' file which lists hosts and the parameters. These relationships could be mapped by dot as a visual path through relationships.

Making a giant leap into the realm of general documentation, in a similar vein, document various pieces of the work flow as separate entities, where the documented entities may be common to several different work flows. Workflows then become linked version of various entities. And dot can generate work flow diagrams.

This all requires organization and effort. In hindsight, I've suggested doing documentation as though you are writing a software program. I'm not sure if tooling exists to do such a thing.

Take it all with a grain of salt. So to speak.

https://news.ycombinator.com/item?id=18674158 is a recent HN on one company's view on documentation.

Maybe automate some of your infrastructure?

I've found that running everything via SaltStatck scripts helps to keep things scripted and documented: as in, the scripts are the documentation (build and maintenance notes are inherently part of the 'process code').

Then apps like dot and similar can be used to build links from the code and attribute files.

That is, attempt to build 'single sources of truth'. And then use tools to 'view' the relationships from their original sources.

By testing the scripts and using the scripts, processes and docs are inherently kept up to date.

Rigorously use the scripts and don't do things manually.

from a programming perspective or a configuration perspective or ?

This reminds me of the story about the elephant in the room.

The article, like most rust articles (in my perception), attacks the 'memory unsafety' issue, and therefore 'everything needs to be converted to rust'.

But ... does rust have the multi-paradigm programming mechanisms which are available in C++?

The article also ignores the issue that the recent releases of C++ are including programming methods for reducing/eliminating the memory issues: RAII, for instance, is a fundamental mechanism for ensuring memory is released at the correct time. Range operators work to eliminate the indexing issues.

'with great power comes great responsibility'. does rust offer the programming tools available in C++? templating, meta-programming, functional coding, procedural programming, object oriented constructs, ....

So, I think the title '.... and Developers Don't Want to Deal with It' is incorrect and click baity. C++ is in active development and evolution, and constructs are coming into place which help to alleviate the issues mentioned in the article.

If I recall correctly, https://www.boost.org/doc/libs/1_68_0/libs/spirit/doc/html/i... uses trie's for it's compile time recursive descent parser for writing grammars and lexical analyzers.

There is a hybrid response here. Those who are self-taught may not be aware of various algorithms or techniques for handling certain scenarios.

Self taught may have fundamental skill sets, but there is benefit in having some sort of educational background which introduces one to some of the more complicated aspects of application design and implementation.

It is that, or being a voracious reader of technical books and being a conference goer, ... ie, basically somehow being able to keep track of and understand the goings-on of current paradigms.

With MQTT, there are a number of broker implementations which offer redundancy/resiliency. I did a quick check and didn't find any place stating anything about any of the routers being able to run in a redundant/resilient mode. If that is available, running WAMP becomes feasible, as I saw some interaction between MQTT and WAMP being possible.

It has an API: https://developers.whmcs.com/api/

What sort of 'seamless integration' issues are you having?

Also, what sort of bookkeeping/accounting system do you use for transaction management?

Maybe not 100%. Bermuda has a root CA: QuoVadis Global.

The document is a late 1980's document, but as such, in trying to say what a NVN structure might look like, they do much to describe existing VN architecture. And it is very telling that not much has changed since then for hardware or software.

Are there running results somewhere? And because of the re-submission ability, are the results any good?

Over the long term, not enjoying coding may become a detriment. Networking is moving more and more towards automation, and all that entails.

Depending upon which world you live, Cisco has their automation, Juniper theirs, and open source has things like Ansible, NAPALM, SaltStack/Proxy,...

Thinking logically and being able to write intent based rules is going to be important.

Do some research on SDN, SDWAN, NFV, ....

Lookup NetFlow as a management tool.

Security, networking, application delivery are all becoming inter-twined and Software Driven.

Adapt or Die.

I run two HP Z32x 4K monitors off an nvidia 1060. Text is easy to see. Lots of pixels. Very little alt-tab. Lots of productivity. So my environment isn't really about retina and HiDPI,

On the other hand, take a look at your developer's postures: are they leaning/squinting into the screen? Are they using 1:1 pixels, or scaling?

What does the retina get you when one is a developer?

Coming back to what I said, I have large monitors so that I can make use of and see the pixels. Any thing smaller would probably be a waste of 4K unless I had really good eye-sight.