Hesitated for a while before posting this, as I don't like to whine in public but this should be therapeutical

Beware, it's a #longread

Years ago, I thought about how cool it'd be to have conversation-based interactive fiction on my phone. I remember showing early prototypes to my ex in 2012. It took me over 2 years to build up the courage to make it my priority and to take time off. FictionBurgers.com was born.

A few weeks in, a friend of mine forwarded me a link to Lifeline. I was devastated. I literally spent 2 days cursing my past self for not making a move sooner.

I soldiered on, worked 7 months straight on it. Now the tech is 90-95% finished, content is maybe 60% finished and I just... gave up. Every other week now, similar projects are popping up. I'm under-staffed and under-financed compared to them. Beyond the entertainment space, "conversation-based" is hot stuff in 2016, and I still can't seem to know what to do with what I have.

I feel like I had this fantastic opportunity and squandered it, which makes me miserable.

Anyway, just so you get some cheese with my whine, here are a few lessons I learned the hard way:

Lesson #1 : Don't go it alone. I thought I could hack it, and for over 7 months, I did. But sooner or later, shit gets to you, it's just human. That's when you need someone; just so that their highs compensate your lows and vice versa. Most of the actual writing was done by a freelancer (and he did AMAZING WORK, especially considering that I couldn't pay him much) but it's not the same as a partner, who's invested same as you.

Lesson #1.5 : Complementary skills. Just like my fiction project failed because I was missing a writer partner, my fallback plan of getting into conversational tech hit the skids for lack of a bizdev partner. It's great to stick among devs when ranting, but you need to mingle with a variety of people. Some of them are actually ok, y'know :)

Lesson #2 : Lean Startup, MVP. Google those terms if you're not familiar with them. My mistake here (after MVPing the shit out of the tech) was to let my content goal run amok : what made my app superior to the competition (or so I reasoned) was that it would allow for conversations with multiple characters! So I started plotting a story... with 9 characters. Not 2 or 3. NINE FREAKING CHARACTERS! Branching conversations with 9 characters is the stuff of nightmare -- and is the main reason I gave up.

Lesson #3 : Know your reasons. I wasted some much time early on, zig-zaging between objectives:
"I'm just indulging myself"
"No, I really want it to be a project that pays off"
"Nah, it's just a learning opportunity"
"Damn, why is it bothering me so much that someone else is doing the same thing ?"
"Doesn't matter, I just mine finished"
"What a waste of time !!"
etc etc
And it's still a problem now that I'm trying to figure out what to do!

So anyway, that's my story, thanks for readin'
Check out chatty.im/player/sugar-wars if you want to test the most advance version.

Also, I've also tagged this #startupfail, if any of you fine people want to share the lessons you've dearly paid to learn!

*team convo*

Dev: We will create another branch for the changes and merge them when required.

Director: Can we not do the changes in the current branch and comment them out until required? I think that’s cleaner and not confusing in my opinion.

Beautiful mf!

Ticket: Allow merchants to customize how their Wallet Passes look! It’ll be super easy, just add these nine merchant-modifiable strings (they support vars) and use their contents for text instead of what we use now. Simple!

Reality: There need to be 24 strings, there are some rules I can’t convey to the merchant (because the system literally does not include instructions, only a name and a textbox), the code to generate the wallet pass is inefficient, uncommented, branching spaghetti that I’ll need to rewrite (it seriously generates every possible field, and then only uses the ones it needs), the specs are so much worse, and half the default values they want aren’t even possible. As in, I don’t know if it’s a car loan, let alone the exact make and model of the bloody thing.

And no, sorry, we have no way of knowing what their fucking “vertical” is, either, so we can’t display that. Fucking sales.

Asdhkjfsjfads

WHY MUST EVERYTHING SUCK

Today on "How the Fuck is Python a Real Language?": Lambda functions and other dumb Python syntax.

Lambda functions are generally passed as callbacks, e.g. "myFunc(a, b, lambda c, d: c + d)". Note that the comma between c and d is somehow on a completely different level than the comma between a and b, even though they're both within the same brackets, because instead of using something like, say, universally agreed-upon grouping symbols to visually group the lambda function arguments together, Python groups them using a reserved keyword on one end, and two little dots on the other end. Like yeah, that's easy to notice among 10 other variable and argument names. But Python couldn't really do any better, because "myFunc(a, b, (c, d): c + d)" would be even less readable and prone to typos given how fucked up Python's use of brackets already is.

And while I'm on the topic of dumb Python syntax, let's look at the switch, um, match statements. For a long time, people behind Python argued that a bunch of elif statements with the same fucking conditions (e.g. x == 1, x == 2, x == 3, ...) are more readable than a standard switch statement, but then in Python 3.10 (released only 1 year ago), they finally came to their senses and added match and case keywords to implement pattern matching. Except they managed to fuck up yet again; instead of a normal "default:" statement, the default statement is denoted by "case _:". Because somehow, everywhere else in the code _ behaves as a normal variable name, but in match statement it instead means "ignore the value in this place". For example, "match myVar:" and "case [first, *rest]:" will behave exactly like "[first, *rest] = myVar" as long as myVar is a list with one or more elements, but "case [_, *rest]:" won't assign the first element from the list to anything, even though "[_, *rest] = myVar" will assign it to _. Because fuck consistency, that's why.

And why the fuck is there no fallthrough? Wouldn't it make perfect sense to write

case ('rgb', r, g, b):
case ('argb', _, r, g, b):
case ('rgba', r, g, b, _):
case ('bgr', b, g, r):
case ('abgr', _, b, g, r):
case ('bgra', b, g, r, _):

and then, you know, handle r, g, and b values in the same fucking block of code? Pretty sure that would be more readable than having to write "handeRGB(r, g, b)" 6 fucking times depending on the input format. Oh, and never mind that Python already has a "break" keyword.

Speaking of the "break" keyword, if you try to use it outside of a loop, you get an error "'break' outside loop". However, there's also the "continue" keyword, and if you try to use it outside of a loop, you get an error "'continue' not properly in loop". Why the fuck are there two completely different error messages for that? Does it mean there exists some weird improper syntax to use "continue" inside of a loop? Or is it just another inconsistent Python bullshit where until Python 3.8 you couldn't use "continue" inside the "finally:" block (but you could always use "break", even though it does essentially the same thing, just branching to a different point).

i was hired to join a team of old devs (40+) in an unnamed European country "yay goodbye 3rd world it's time to enjoy the quality of life" assist with enhancing already existing software and creating new solutions.
prior to my arrival most things were slow and super buggy, looking at the code base it shouldn't be a surprise, amateur hour everyone, logic implemented that is not needed, comment driven development, last time code review was done back in 1996. lots of anti patterns.
i swear there is a for loop that does nothing but it loops through a 100+ elements list, trunk based development with tfs since git is "not really needed"
test projects are not there.
>enter me an educated fool, with genuine passion for the craft and somehow a decent amount of knowledge.
>spent the last year fixing stuff educating people on principles and qualities.
> countless hours of training and explaining. team is showing cooperation, a new requirement comes in to develop with react.
> tear my ass creating reusable shit and self explanatory code with proper naming etc using git with feature branching, monday is first deployment day.
> today a colleague was working on an item submit a pull request and self approve it
> look at the code..... WTF the dumb fuck copied and pasted the whole code from different kendo components but somehow managed to refractor the name to test component, commented out all the code that he didn't use did the api call directly from the component, has 2 useeffects that depends on the a fucking text box changes for no reason, no redux implementation, the acceptance criteria is not achieved, and it doesn't work it just look right.
> first world country shit cannot scold, cannot complain, lead by example.
>asked him why you did this, the response was yeah probably i shouldn't have done that, i really didn't understand anything in the training but didn't want to waste time!!!!
> rest of the team created a different styled disaster with different flavors they don't even name their shit the same way.

fellow developers I'm stuck in a spaceship with a bunch of imposters, seriously i never cried in my entire life now I'm teary and on the verge of a break down.
talk with management "improving needs time" and offers me to join a yoga session to release the stress as if reaching nirvana would deliver shit on monday.
i really don't know what do is this a rant, is this a cry for help, I'm not sure, any advice is welcomed.

A newly joined developer (who was supposed to be very senior) comes and asks me how to write a test cos for some reason the person didn't know how to mock.

In Java,
(same for any other implementation which has an interface)
Writes Arraylist list =.....
Instead of List list = Arraylist...

Deployed code (another engineer from another country helped to deploy since this new senior dev didn't have access yet.

But the new senior dev didn't update relevant files in production code which brought down the site for nearly an hour. Mistake aside, the first reaction from this new senior dev is 'WHY DIDN'T THE DEV THAT WAS HELPING DIDN'T DO THE FILE UPDATE?'

This was followed by some other complaints such as our branching stragies are wrong. When in fact the new senior dev made a mistake by just making assumptions on our git branching strategies and we already advised on correct process.

Out of all these, guess this is the best part. The senior dev never tested code locally! Just wrote code, unit test and send to QA and somehow the test passed through. I learnt this when I realised this dev... has not even set up the local environment yet.

I keep saying new but this Senior dev been around like 3 months! This person is in another team within our larger team but shares same code base. I am puzzled how do you not set up your environment for 3 months. Don't you ask for help if you are stuck? I am pretty sure the env is still not setup.

Am I over reacting or is this one disgusting developer who doesn't even qualify for an intern let alone a senior dev? It's so revolting I can't even bring myself to offer help.

I use version control as a glorified backup. Only recently did we start branching at work.

This is why I need to be part of a proper team where I can learn instead of being a team of 2 juniors and no one else 😂

... worst drunk coding experience?
none. or to be more precise, all of the three of them I had. I can't code drunk, i hate doing it, i hatw even thinking about doing it when drunk.

so after those initial three attempts i don't try to do it again, ever.

BUT, best coding experience while high?
ALL OF THEM.
some of the best pieces of code I wrote i did when I was high. my mind goes into overdrive at those times, and my thinking is not lines/threads of thought, but TREES of thought, branching and branching, all nodes of each layer of the tree coming to me AT ONCE, one packet == whole layer across all of the branches.

and the best was when one day, in about 14 hour marathon of coding while high, i wrote from scratch a whole vertical slice of my AI system that i've been toying around in my head for several years prior, and I had all of the high-level concepts ALMOST down, but could never specify them into concrete implementations.

and I do mean MY ai system, my own design, from the ground up, mixing principles of neural networks and neuropsychology/human brain that I still haven't seen even mentioned anywhere.

autonomous game ai which percieves and explores its environment and tools within it via code reflection, remembers and learns, uses tools, makes decisions for itself for its own well-being.

in the end, i had a testbed with person, zombie and shotgun.
all they had pre-defined in their brains were concepts of hunger and health. nothing more.

upon launching it, zombie realized it wants to feed, approached oblivious person, and started eating it.

at which point, purely out of how the system worked, person realized: "this hurts, the hurt is caused by zombie, therefore i hate zombie, therefore i want to hurt it", then looked around, saw the shotgun, inspected its class by reflection, realized "this can hurt stuff", picked the shotgun up, and shot the zombie.

remembered all of that, and upon seeing another zombie, shot it immediately.

it was a complete system, all it needed to become full-fledged thing was adding more concepts and usable objects, and it would automatically be able to create complex multi-stage, multi-element plans to achieve its goals/needs/wants and execute them. and the system was designed in such a way that by just adding a dictionary of natural language words for the concept objects on top of it, it should have been able to generate (crude but functional) english sentences to "talk" about its memories, explain what happened when, how it reacted, what it did and why, just by exploring the memory graph the same way as when it was doing its decision process... and by reversing the function, it should have been able to recieve (crude) english sentences that would make it learn what happened somewhere else in the gameworld to someone else, how to use stuff and tell it what to do, as in, actually transfer actual actionable usable knowledge to it...

it felt amazing to code for 14 hours straight, with no testruns during that, run it for the first time after those 14 hours, and see that happen.

and it did, i swear! while i was coding, i was routinely just realizing typos and mistakes i did 5-20 minutes ago, 4 files/classes ago! the kind you (and i) usually notice only when you try to run the thing and it bugs out.

it was a transcendental experience.

and then, two days later, i don't remember anymore what happened, but i lost all of that code.

and since then, i never mustered enough strength and resolve to try and write the whole thing again.

... that was like 4 years ago.
i hope that miracle will happen again one day...

Working with a data scientist on an update to a machine learning api that has dinner logic change with a new model.

He's wondering why his PRs are falling.

He's trying to merge into development from a branch created off of main.

He's renamed all the functions and classes and never updated the call points.

He's using new packages but never includes them in the requirements file, so the docket builds are failing.

His class method definitions don't contain self and are throwing syntax errors.

I've been working with him for 4 days to get him to understand branching, linting, unit testing, and not blindly copy and pasting snippets from jupyter notebook into production api code!

Alright, my very first post here was about this project and I am thinking it out loud again.

I see a problem and I am struggling to find a solution.

Now what I am thinking of is to articulate the problem well and state WHY I believe it needs to be solved. There are some reasons which must be presented in a capitalist way.

Furthermore, I am thinking of doing a market research to understand various demographics, validate the idea, and figure out the product-market fit.

Now, this qualitative research and quantitative data will help me decide whether it is worth putting in the efforts to solve the problem or not.

And since, we have an MVP already (funnily yes, we built it before all of the above), that will help me validate the tangible solution.

Once we get a confidence boost, then it will be time to get that single transaction which has net positive cash flow.

Start scaling to 'next billion users', so a billion transaction with net positive cash flow.

I won't be branching out into multiple verticals before be able to sustainably scale the core USP.

And while the second half sounds like, 'I have a million dollar idea', I am trying to be more and more realistic and rationale instead of falling in love with my idea.

I don't even have an idea (read solution) to fall in love with. Rather I have a problem that is bothering me.

So, yes, I am continuing this journey to solve the problem which started in second year of my hostel room and has evolved over 10 years.

My work product: Or why I learned to get twitchy around Java...

I maintain a Java based test system, that tests a raster image processor. The client is a Java swing project that contains CORBA bindings to the internal API of the raster image processor. It also has custom written UI elements and duplicated functionality that became available in later versions of Java, but because some of the third party tools we use don't work with later versions of Java for some reason, it's not possible to upgrade Java to gain things as simple as recursive directory deletion, yes the version of Java we have to use does not support something as simple as that and custom code had to be written to support it.

Because of the requirement to build the API bindings along with the client the whole application must be built with the raster image processor build chain, which is a heavily customised jam build system. So an ant task calls out to execute a jam task and jam does about 90% of the heavy lifting.

In addition to the Java code there's code for interpreting PostScript files, as these can be used to alter the behaviour of the raster image processor during testing.

As if that weren't enough, there's a beanshell interface to allow users to script the test system, but none of the users know Java well enough to feel confident writing interpreted Java scripts (and that's too close to JavaScript for my comfort). I once tried swapping this out for the Rhino JavaScript interpreter and got all the verbal support in the world but no developer time to design an API that'd work for all the departments.

The server isn't much better though. It's a tomcat based application that was written by someone who had never built a tomcat application before, or any web application for that matter and uses raw SQL strings instead of an orm, it doesn't use MVC in any way, and insane amount of functionality is dumped into the jsp files.

It too interacts with a raster image processor to create difference masks of the output, running PostScript as needed. It spawns off multiple threads and can spend days processing hundreds of gigabytes of image output (depending on the size of the tests).

We're stuck on Tomcat seven because we can't upgrade beyond Java 6, which brings a whole manner of security issues, but that eager little Java updated will break the tool chain if it gets its way.

Between these two components we have the Java RMI server (sometimes) working to help generate image data on the client side before all images are pulled across a UNC network path onto the server that processes test jobs (in PDF format), by reading into the xref table of said PDF, finding the embedded image data (for our server consumed test files are just flate encoded TIFF files wrapped around just enough PDF to make them valid) and uses a tool to create a difference mask of two images.

This tool is very error prone, it can't difference images of different sizes, colour spaces, orientations or pixel depths, but it's the best we have.

The tool is installed in both the client and server if the client can generate images it'll query from the server which ones it needs to and if it can't the server will use the tool itself.

Our shells have custom profiles for linking to a whole manner of third party tools and libraries, including a link to visual studio 2005 (more indirectly related build dependencies), the whole profile has to ensure that absolutely no operating system pollution gets into the shell, most of our apps are installed in our home directories and we have to ensure our paths are correct for every single application we add.

And... Fucking and!

Most of the tools are stored as source bundles in a version control system... Not got or mercurial, not perforce or svn, not even CVS... They use a custom built version control system that is built on top of RCS, it keeps a central database of locked files (using soft and hard locks along with write protecting the files in the file system) to ensure users can't get merge conflicts by preventing other users from writing to the files at all.

Branching is heavy weight and can take the best part of a day to create a new branch and populate the history.

Gathering the tools alone to build the Dev environment to build my project takes the best part of a week.

What should be a joy come hardware refresh year becomes a curse ("Well fuck, now I loose a week spending it setting up the Dev environment on ANOTHER machine").

Needless to say, I enjoy NOT working with Java. A lot of this isn't Javas fault, but there's a lot of things that Java (specifically the Java 6 version we're stuck on) does not make easy.

This is why I prefer to build my web apps in python or node, hell, I'd even take Lua... Just... Compiling web pages into executable Java classes, why? I mean I understand the implementation of how this happens, but why did my predecessor have to choose this? Why?

What branching methodology do you use and why? We've been using a trunk based development model, but I'm reviewing others.

Questions more then a rant...

I've moved from being a lead on imploring DevOps and Agile practices in a large Telco to now working for a security consultancy... The team I'm with are s*** hot when it comes to SecOps (which is why I changed jobs) and I've been hired to he the automation and working practice expert on the team. Already got some of them learning Ansible which is a great start!

I've got delivery now being pushed to Git and all client work being tracked in Jira and properly documented and collaborated through HipChat and other CI tools on the way....

My question is this... Does anyone have some awesome resources to teach people Git, Jira, Jenkins, etc. quickly without forking or branching out on expensive training? Focus on being a technical but consultative team. Ideally just wanna pull some awesome guides and make. My own commits on them for the team... Please fire a story or epic away!