Once a project starts going non-binary that's when I know they've really lost it.

It's a spectrum of numbers, you hetero-bit-normative supremacist.

If I remember right, Lenovo, Dell, HP, and some others had their own recovery partitions at some point so that if you used a recovery disc of theirs it had instructions to reload all the bloatware and call it a "clean default install". I see various Lenovo shortcuts in the background. There is a possibility of that being partly to blame.

Yes, the whole group at the same time.

Break that sheep (I hate using that term, but it fits) mindset. Think of yourself as an one-man company and you are interviewing potential partnerships. Keep that thought driving your actions, you are interviewing them just as much as they are interviewing you.

Probe for social dynamics, you can do it indirectly and subtly by having everyone in one room talking together. Ask what was the reason for choosing tools and processes, you just want to know why so you can get a better idea of how things work at the company. Inquire about how the company is hoping you can improve things, this sets you apart from code monkies.

Find smaller companies. They can still have people that cause trouble, but it is a little easier to catch that in an interview. Ask to be seen by the whole group (you would work with) for 15 or 20 minutes to find out about everyone, because you want to be sure of that position being a good fit. Ask them to tell you about working at that company and themselves. Who does what and who works with whom. Pay attention to who sits where and who feeds off of whom (do people make jokes and involve each other, or is everyone tight-lipped with a stick up their butt?). Who constantly chimes in, how relevant is their comment/response, and how much did they actually say when answering your questions. Sometimes people can speak for a while without actually having anything to say. It will be a bit mentally exhausting when you do it the first few times but it can be a wealth of information.

I've sat in on a meeting or interview and spent half of my mental energy on taking mental notes of how people were sitting, it let me know who was really in charge and how everyone was feeling. I've seen meetings where a small team of programmers came in and everyone was hunched over and quiet, a giant red flag they were all mentally roughed up and needed a break.

People advertise their thoughts all the time amd don't realize it, if you are actively looking for the signs they are easy enough to pick up on. From these silent messages you can tell if the place will be worth your time.

It looks like there is a sort of progress meter based on what monitor has a green screen, I think.

I gave an answer once about joining tables in MySQL. I got a comment about how I was completely wrong and had no clue what I was talking about. I responded that he should have a talk with my MySQL server because it was obviously also doing things wrong by working with the SQL statement I used. Don't think he responded to that. Ever since then I've felt like StackOverflow is a cancer that I try to avoid.

The use case for me is an e-commerce situation where we want to create a copy of the user's cart upon checkout, but we don't want to leave entries in a cart table that is constantly growing. We'd rather deal with it on the same table we track orders with so when we archive them we have all the relevant information in a single record, not related records. Having a JSON column to store the cart data is a whole lot easier than who knows how many pivot table entries.

How well does 10.2 handle JSON data? That's been one thing that I could use that MariaDB has failed on me in the recent past while testing it.

That's great that you feel that way, it doesn't mean much to me because I'm not, nor do I want to be, filled in on the context of your situation. My response was simply trying to point out that it can go both ways. Processes/methodologies can help or they can hurt, I won't argue that point. The point of the article I linked that I felt was relevant to most any situation in programming is the idea that it is imperative to think through the code before writing it. That's it. A sign that code wasn't thought through or written well is when the excuse of "it's too complicated to test" is used.

Everything you described is all about the people, not the process. The process can't hurt anyone, it can't fire anyone, it can't yell at anyone.

Just because someone can break a design guide doesn't mean they should. I've seen people who didn't care at all about design guides and it was a nothing but a tangled mess. Code that was so convoluted because no one stopped to think things through. Once I point out something to them about how their process could be so much more simple and linear, all I get is a soft "oh, well I guess that could work".

I'd also like to point out that the article didn't talk about Agile or Scrum, it only talked about TDD.

One part of the article that I felt was poignant:

Any design that is hard to test is crap. Pure crap. Why? Because if it’s hard to test, you aren’t going to test it well enough. And if you don’t test it well enough, it’s not going to work when you need it to work. And if it doesn’t work when you need it to work the design is crap.

When the author refers to designing:

You have to DESIGN period. No matter what you are writing; whether a unit test, or an acceptance test, or production code, or a mock, or a stub, you have to DESIGN.

It does have some good links (that I scanned through) but I have to be honest, I am not a fan of the idea/push for a "full-stack developer" as a single position in a company. I am supportive of knowing how to do a little on one end or the other* but the idea that one person can really know enough about either side (front or back) to be properly effective is something that doesn't sit right with me. I think it takes enough time and energy to become proficient in one that it can take away from the other as you try to keep updated on current trends and changes. Again, I think it's really good to know different languages especially ones that you use in different scenarios**. I feel doing so can open up ideas and understanding about the general practice of programming/problem solving. I prefer taking the "Ocean's 11" approach and getting a good team that works well together but each individual (or pairs, whatever) that focus in a specific section.

*e.g. a back-end developer knowing what build tools are and how to use one or two along with knowing what is current for front-end developers

**e.g. C++ on a system service or desktop situation, Python for web development, JavaScript/Dart/whatever for client-side stuff

The free version is actually pretty good on features, the biggest perk I see from the paid version is import from one and export to another (MySQL to MSSQL). Really cool to see the Linux compatibility.

• Fedora Desktop
• Sublime Text 3
• Nginx+PHP-FPM
  • PHP-FPM is running as my user, code goes in a directory in my home directory
  • sessions, cache, and temp directories for PHP are reside in a dot directory in my home directory
• Valentina Studio for any SQL work I don't want to do on a CLI

Are you using command-line or a desktop application?

A big influence on getting a programming job (or tech in general) is knowing someone and getting a job with their employer. Referrals can mean a lot.

I've gotten a job by showing up dressed well and answering, "No, sir." when asked if I'm lazy. That's all it took because I knew someone.

Good on you for at least starting it and seeing how easy it is to get going.

I've been using Hacker Rank to pick up the basics of C++. I like it for the simple fact that I can try out something without having to install everything to my computer to see if I can do the basics well enough to invest serious time and energy in one language.

/*
You will THINK you can improve this method without re-writing that nasty part of the legacy code base.
You will ATTEMPT to improve this method.
You will FAIL to do it after countless hours.
You have been warned.
*/

Oh dear heavens, you seemed so smart and I had high hopes for you. It seems you haven't seen the lights of the Vim Vanguard.

I'm before PHP is brought up.

But seriously listen to SamBone, he just laid out some great wisdom.

Wait, this can't be right. This looks like a well reasoned response. We can't have that kind of thinking here! This is /v/programming! We have more holy wars than the Middle East.

I was in middle school and I used BASIC(?) to write simple applications on my graphing calculator to make math easier. I never understood why we needed to memorize some formula and then do 30 problems with it. Since the formulas are using using variables, I prompted for the values, printed out the steps with the real values and the eventual answer.

First math teacher I had that found out asked me not to share the code because if I could write it, then I understood the math. Another teacher later on said I was cheating and threatened to take my calculator away. I said it wasn't fair to do that because everyone had the same tools as I did and I shouldn't be punished for using them better than others.

I usually go with the common denominator for a language. E.g. PHP has PSR-2.

A space is a space, no matter the editor. Most editors have a "translate tabs to space" feature so you can use tabs like you normally would.

I use spaces, one tab is four spaces.

Edited to add: use whichever you want but for the love of all that is good and holy, be consistent. I convinced my team at work that it was important to all choose one or the other and to format any file correctly if they touched a single line in it. At first it meant a lot of big merge requests, but over a two month period every one started to say the code was more readable because of it. I was desperate for consistency I even offered to do tabs even though I am a vocal spaces supporter.

I also really hate the attitude towards new users, as if the first thing you do when starting to program is make an account with them so if your account is new you must be new to programming. It also seems to have a perception that there is a constant accumulation of points over time, so if your account is past a certain age you should have a set amount of points or you must be stupid.

I still remember posting once about an SQL question, and while my answer wasn't the best it was one that I was using on a personal site at the time. I knew it worked but I still got the asinine comment of "this is awful and won't work". No explanation as to why it shouldn't work, or how to improve it, just that it was wrong. I sarcastically replied that he should definitely have a sit down with my server then because it was obviously going outside the bounds of it's programming by working for me.

I started in tech support and moved into coding websites nearly a decade ago. One of the best ways to build up a network of people is to go to local meetups or conferences. If you are lucky enough to live in an area that has moderate sized conferences usually you can volunteer to be a gopher for them and then get in for free. You spend a few hours just running around, take a cable to this presenter, double check that the camera is recording, ensure the presenter has their mic on and it is recording, bring water to that presenter, help people check in and get their badges in the morning. Then you spend time during the lunch/dinner socials just meeting people. Exchange email addresses, follow them on Twitter, connect on LinkedIn, whatever they have. Don't go crazy asking every person who says "hi" for this information, but if they do strike you as someone you want around you then do so.

What language(s) are you looking to learn? Did you also want DevOps experience (setting up a LAMP/LEMP server)?

Now that I think about it, the coworker sounds like an intelligent and handsome chap that I'd like to meet.

Oh, no worries but nope.

I don't get it.

That is another good reason to use Agile, but like you pointed out it comes down to good people doing it right.

Something I noticed about your post was that it offered zero alternatives to any of the problems you described. If all you want to do is rant, then ignore the rest of this comment because it won't be relevant.

Agile methodologies are really centered around one idea: figure out exactly what problem(s) you are trying to solve before you begin writing code. Yes, many people go nuts with the whole process to the point of creating jobs out of the concept. However, I've seen projects without it (and unit tests) fail horrendously. Some with those fail too, which means they don't seem to be the problem.

Unit tests are just having an objective way to test functionality. I've worked at places that fought against unit testing and invariably someone changes some core functionality because it was written terribly. It was written terribly because that's how the lead developer learned how to do things 10 years prior. Things break across the board, people get mad because the person who changed the code didn't test everything. If I spent the time to test every endpoint in an API more than half of my day would be consumed with testing. So of course I layer my voice with sarcasm when I say, "Gee, if only there were some way to AUTOMATE checking if this change broke things. You know? Like, it could TEST it all for us. Even go through each method, or UNIT, of the application. Golly, that would sure be helpful!"

After the angry looks come my way I just tell them this is the exact kind of problem unit testing is meant to prevent. You don't need absolute code coverage, but it can be highly effective to cover the core components. And tests failing are not a bad thing, pushing out code that broke the tests is. No one gets it right the first time, and if your developers are too afraid to do anything because it breaks a test then the problem is with the culture/environment.

As for rapid release, if it all works then there should be no problems when or how often a release comes. Again, all the things you are bringing up sound like people problems, not methodology problems.

Very cool!

Are there any local user groups? They can be a good mine.

Honestly I can't name specific companies because it depends on where you are and your skill set. I live in Utah, and there is a rather large tech company presence here. It seems that many of the smaller ones can't hire juniors because they are running "lean", but the larger ones tend to be able to take on juniors.

My advice to finding companies that hire juniors is find a local user group for your programming language of choice with an email list and send out an email asking if anyone knows which companies are junior friendly. Odds are you'll get a handful of responses, some may even tell you why their particular company is great for juniors.

What do you look for when identifying the diamonds in the rough?

I've usually found that those types are ones to think through problems and don't stop the first time it works, they make sure to account for user error, odd cases, etc.

I haven't looked at it in a long time, but when I was in college I liked CodeBlocks.

While I haven't written anything with it, I have been using Go Got Service/Gogs and I like it. It has been fast for me on a small VPS and uses well under 100Mb of memory. I have tried GitLab, and while I liked the features it required too much in the way of resources to make running a small private git server financially worth it.

They may have been structured for cross-platform builds, but it should have been filtered to have Linux in each one.

I'm not a C++ developer, but your question made me curious about project directory structure. I just looked at trending C++ projects on GitHub and thought the RethinkDB project seemed like it might a good starting point. Hope that helps.

Another project that might be of some indication, but seemed completely different, was the Vulkan samples project.

I'm really sorry I don't have your answer, but I wanted to say you did an awesome job asking your question.

Way to stereotype. I'm sure everything you just posted could be used to get you thrown out of any online community because you just made programmers feel unsafe.

But then again, I'm an adult and can shrug off your opinion. Which is why I'm against CoCs, because I expect everyone to just be an adult and act appropriately.

I love seeing things like this. It used to be that you had to constuct a small wall of divs wrapping the one thing you wanted to target and then go on a JavaScript binge to get it to behave the way you wanted.

I think JavaScript has incredible uses but for a while it was getting abused which left a lot of front-end devs considering CSS for just layout, static designs, and typography. It wasn't until some of the browser devs started to take all the stuff in CSS that JavaScript was manipulating and make it easier to deal with in CSS that we started to see stuff like this.

That looks really intriguing. The color scheme reminds me a bit of the older Ubuntu Gnome 2 default theme, but still a really cool idea.

I've had that happen a few times. I look at a word and all I can think is, "I know it's spelled right but the word looks absolutely wrong".

I couldn't agree more.

I think it happens to almost every developer at one point. One thing that I've found to be of some help is to listen to Amy Hoy advice on bootstrapping a project. Not that you have to launch products, but a lot of her free newsletters contained an idea or two that really hit home with me.

Some of her advice that may help you is to take one of your unfinished projects and look at it from the beginning. Who is the target audience, what purpose will it serve for them, and what is the minimum you need to deliver? If you can't do everything already for what the minimum requires, list everything that is a requirement and review them to see if it is really necessary and if it is, what exactly are the steps needed to accomplish it. As this process goes on, you'll have a good plan of what you need to do and what you need to learn. Keep breaking down what you need to learn and work on that one small piece at a time. Build on little wins and progress.

I understand that trade-off, you get a free product and they get data on how it's being used to improve it. I'm completely fine with that sort of thing if it is clear from the beginning. I hate having to read terms of service to find out how much data will be collected on me. If I'm paying for a product then I expect that privacy, if it's free I assume data collection will happen but I like to think it won't be too intrusive. But when I read things like this:

We collect this data in a variety of ways, including from web forms and software running on your device. We also obtain data from other sources and combine with data we collect directly.

I don't like how vague it is, "other sources" could be practically anything.

Thankfully, in this case, Microsoft didn't have a gigantic legal wall of text. It was rather clear and concise. So some points to them for not burying it in legal jargon.

Not sure I'm up for switching from Sublime Text. From their License:

The software may collect information about you and your use of the software, and send that to Microsoft. Microsoft may use this information to provide services and improve our products and services. For this pre-release version, users cannot opt out of data collection. There may also be some features in the software that enable collection of data from users of your applications. If you use these features to enable data collection in your applications, you must comply with applicable law, including providing appropriate notices to users of your applications.

And from their Privacy Statement:

Microsoft collects data to operate effectively and provide you with the best products, services and experiences we can. Some of this information you provide directly to us, such as when you register for a service like Visual Studio Online. Some of it we get by recording how you interact with our products and services. We collect this data in a variety of ways, including from web forms and software running on your device. We also obtain data from other sources and combine with data we collect directly.

[Edited to add] Full disclosure: I have Sublime Text. I happily paid for it after using an unregistered copy for about a month. Sublime's EULA only goes over it being copyright work, what computers you can use it on, and that they aren't held liable if it crashes and you didn't save for 3 hours. I have no qualms with paying $70 USD for my privacy.

To add to this I'd say make morning and evening routines. Usually what I do in the morning:

1. As soon as I get out of bed I pray then meditate. The purpose of the prayer is to think of all the ways I have it good. I have people who love me, I earn enough to provide for my family, I do something I enjoy every day. I then take a few minutes to clear my head, think about nothing. Then I meditate. It's really just holding still in a different position where I focus on what I want to accomplish for the day, in the order I want to do it. Think about what it will take to accomplish it all. But don't get hung up if I don't accomplish all of it.

2. Eat a good, healthy, breakfast. Lately it's been raisin bran with banana cut up in it. Find whatever works for you, but make it good for you. Try to get in proteins, eggs are a good source.

3. Shower and get dressed in clean clothes. This is important for me because I work from home. If I'm in the mindset of "going" to work, then I am far more productive than if I hang out in what I slept in.

4. Take 30 minutes in the morning to learn something unrelated. Chess Academy is what I've liked in the past. This kind of activity helps you warm up your brain.

Then I actually get to work. The days I stick to my routine are orders of magnitude better than when I don't.

In the evenings, at a given time, I brush my teeth, wash my face, then lay day in bed. To get myself to wind down I think about my day and what I accomplished. Reflect on what I meditated on that morning. How much did I accomplish, did I not get it all because of external forces or did I screw it up by not being disciplined enough. Focus on the good, and it usually helps me do better the next day.

While it's not an IDE, I really like SublimeText.

I noticed this is just the syntax highlighting packages, but still really cool that this move was made. I think it will take a huge load off Jon Skinner's shoulders and hopefully keep that functionality of SublimeText at the level it is now.

It's a great question but there is a caveat to it: what platform are you aiming at?

Which of the following applies:

• desktop OS application
• desktop website (non-mobile interface)
• mobile website (geared specifically at low-bandwidth and small screen)
• mobile application (native)
• mobile application (non-native, e.g. uses HTML/CSS/JS instead of Objective C/Java)

Here are my two credits worth of thoughts on each.

On the desktop I'd go for a calculator. It can be as simple as basic arithmetic to more advanced features like graphing output on a formula using a GUI library.

A website, either desktop or mobile, do a blog. Not just any blog, but one written using object-oriented code. Do role-based permissions, repository patterns, events, JSON (or XML) API, handling uploads, etc. Can be done with or without a framework.

Mobile apps I'm not really well versed on a concept that would be great to cover basics of each. For that, someone else will have to chime in.

Oh they already did that for the HTML/CSS. Tables. Tables everywhere. Nested. I was surprised it didn't say "best viewed in IE".

Same kind of mentality I had to face. It was salt in the wound because I took the job out of necessity. I had to drive over 40 miles one way, so it took me nearly an hour because of traffic when driving.

I can't remember it exactly, but it was something like:

Don't edit this function. It looks screwy, you know It, I know It.
But it works. Seriously, don't change it. $TeamLead, I'm talking to you.

Backstory: My team lead was heavily against using any 3rd party code, including frameworks (writing PHP like it was 1998, but it was just two years ago!).

So I had to writing something that does pretty much what Carbon does, but object oriented code wasn't used on this project. Had to do it as procedural with several functions. Any attempt to just scrap the file and rewrite it would end up just as corrupted with spaghetti as before as he made some edits. I'd get it working, he'd break it again. I lasted 6 months at that job.

Your point about who will be doing more work may have the most pull.

Disclosure: I use Laravel for my day job.

You'll also want to take into account the target user base. What's is important, fast prototype? Rapid change implementation? How much server resources will it take to set up a server to run Java? I ask because I don't know, I'm not being sarcastic. How quickly do you want change features? I've found that Java can take longer than I'd like because of build times depending on what you're doing. But the trade off is once it's built, it's built.

Java can shine when it comes to serving speed, but server resources are a factor. Laravel is solid and composer packages give amazing options for functionality. Based on what you said the project would involve, sounds like both could work. I'd opt for Laravel because of the features you listed seem to be the exact sort of thing PHP/Laravel was built for.

Again, just my opinion but I'd like to hear from someone with Java experience chime in on what kind of server requirements you'd be looking at.

Do you consider web development (mostly PHP but some other stuff too) part of that criteria? If so then yes, I have no degree and I've been working for 5 years.

I really like all of Blackmill's work. It is more mellow but I've gotten into some serious coding grooves while listening to his stuff.

I also listen to audio books that I already know, usually I've read the book before. It gives me something to have as idle thoughts while I'm focusing on a problem. Sometimes I have to turn it off to really focus, but that's not too often.