I’m surrounded by idiots.

I’m continually reminded of that fact, but today I found something that really drives that point home.

Gather ‘round, everybody, it’s story time!

While working on a slow query ticket, I perused the code, finding several causes, and decided to run git blame on the files to see what dummy authored the mental diarrhea currently befouling my screen. As it turns out, the entire feature was written by mister legendary Apple golden boy “Finder’s Keeper” dev himself.

To give you the full scope of this mess, let me start at the frontend and work my way backward.

He wrote a javascript method that tracks whatever row was/is under the mouse in a table and dynamically removes/adds a “.row_selected” class on it. At least the js uses events (jQuery…) instead of a `setTimeout()` so it could be worse. But still, has he never heard of :hover? The function literally does nothing else, and the `selectedRow` var he stores the element reference in isn’t used elsewhere.

This function allows the user to better see the rows in the API Calls table, for which there is a also search feature — the very thing I’m tasked with fixing.

It’s worth noting that above the search feature are two inputs for a date range, with some helpful links like “last week” and “last month” … and “All”. It’s also worth noting that this table is for displaying search results of all the API requests and their responses for a given merchant… this table is enormous.

This search field for this table queries the backend on every character the user types. There’s no debouncing, no submit event, etc., so it triggers on every keystroke. The actual request runs through a layer of abstraction to parse out and log the user-entered date range, figure out where the request came from, and to map out some column names or add additional ones. It also does some hard to follow (and amazingly not injectable) orm condition building. It’s a mess of functional ugly.

The important columns in the table this query ultimately searches are not indexed, despite it only looking for “create_order” records — the largest of twenty-some types in the table. It also uses partial text matching (again: on. every. single. keystroke.) across two varchar(255)s that only ever hold <16 chars — and of which users only ever care about one at a time. After all of this, it filters the results based on some uncommented regexes, and worst of all: instead of fetching only one page’s worth of results like you’d expect, it fetches all of them at once and then discards what isn’t included by the paginator. So not only is this a guaranteed full table scan with partial text matching for every query (over millions to hundreds of millions of records), it’s that same full table scan for every single keystroke while the user types, and all but 25 records (user-selectable) get discarded — and then requeried when the user looks at the next page of results.

What the bloody fucking hell? I’d swear this idiot is an intern, but his code does (amazingly) actually work.

No wonder this search field nearly crashed one of the servers when someone actually tried using it.

Asdfajsdfk.

I had to open the desktop app to write this because I could never write a rant this long on the app.

This will be a well-informed rebuttal to the "arrays start at 1 in Lua" complaint. If you have ever said or thought that, I guarantee you will learn a lot from this rant and probably enjoy it quite a bit as well.

Just a tiny bit of background information on me: I have a very intimate understanding of Lua and its c API. I have used this language for years and love it dearly.

[START RANT]

"arrays start at 1 in Lua" is factually incorrect because Lua does not have arrays. From their documentation, section 11.1 ("Arrays"), "We implement arrays in Lua simply by indexing tables with integers."

From chapter 2 of the Lua docs, we know there are only 8 types of data in Lua: nil, boolean, number, string, userdata, function, thread, and table

The only unfamiliar thing here might be userdata. "A userdatum offers a raw memory area with no predefined operations in Lua" (section 26.1). Essentially, it's for the API to interact with Lua scripts. The point is, this isn't a fancy term for array.

The misinformation comes from the table type. Let's first explore, at a low level, what an array is. An array, in programming, is a collection of data items all in a line in memory (The OS may not actually put them in a line, but they act as if they are). In most syntaxes, you access an array element similar to:

array[index]

Let's look at c, so we have some solid reference. "array" would be the name of the array, but what it really does is keep track of the starting location in memory of the array. Memory in computers acts like a number. In a very basic sense, the first sector of your RAM is memory location (referred to as an address) 0. "array" would be, for example, address 543745. This is where your data starts. Arrays can only be made up of one type, this is so that each element in that array is EXACTLY the same size. So, this is how indexing an array works. If you know where your array starts, and you know how large each element is, you can find the 6th element by starting at the start of they array and adding 6 times the size of the data in that array.

Tables are incredibly different. The elements of a table are NOT in a line in memory; they're all over the place depending on when you created them (and a lot of other things). Therefore, an array-style index is useless, because you cannot apply the above formula. In the case of a table, you need to perform a lookup: search through all of the elements in the table to find the right one. In Lua, you can do:

a = {1, 5, 9};
a["hello_world"] = "whatever";

a is a table with the length of 4 (the 4th element is "hello_world" with value "whatever"), but a[4] is nil because even though there are 4 items in the table, it looks for something "named" 4, not the 4th element of the table.

This is the difference between indexing and lookups. But you may say,

"Algo! If I do this:

a = {"first", "second", "third"};
print(a[1]);

...then "first" appears in my console!"

Yes, that's correct, in terms of computer science. Lua, because it is a nice language, makes keys in tables optional by automatically giving them an integer value key. This starts at 1. Why? Lets look at that formula for arrays again:

Given array "arr", size of data type "sz", and index "i", find the desired element ("el"):

el = arr + (sz * i)

This NEEDS to start at 0 and not 1 because otherwise, "sz" would always be added to the start address of the array and the first element would ALWAYS be skipped. But in tables, this is not the case, because tables do not have a defined data type size, and this formula is never used. This is why actual arrays are incredibly performant no matter the size, and the larger a table gets, the slower it is.

That felt good to get off my chest. Yes, Lua could start the auto-key at 0, but that might confuse people into thinking tables are arrays... well, I guess there's no avoiding that either way.

I wrote a database migration to add a column to a table and populated that column upon record creation.
But the code is so freaking convoluted that it took me four days of clawing my eyes out to manage this.
BUT IT'S FINALLY DONE.
FREAKING YAY.

Why so long, you ask? Just how convoluted could this possibly be? Follow my lead ~

There's an API to create a gift. (Possibly more; I have no bloody clue.)
I needed the mobile dev contractor to tell me which APIs he uses because there are lots of unused ones, and no reasoning to their naming, nor comments telling me what they do.

This API takes the supplied gift params, cherry-picks a few bits of useful data out (by passing both hashes by reference to several methods), replaces a couple of them with lookups / class instances (more pass-by-reference nonsense). After all of this, it logs the resulting (and very different) mess, and happily declares it the original supplied params. Utterly useless for basically everything, and so very wrong.

It then uses this data to call GiftSale#create, which returns an instance of GiftSale (that's actually a Gift; more on that soon).

GiftSale inherits from Gift, and redefines three of its methods.

GiftSale#create performs a lot of validations / data massaging, some by reference, some not. It uses `super` to call Gift#create which actually maps to the constructor Gift#initialize.

Gift#initialize calls Gift#pre_init (passing the data by reference again), which does nothing and returns null. But remember: GiftSale inherits from Gift, meaning GiftSale#pre_init supersedes Gift#pre_init, so that one is called instead. GiftSale#pre_init returns a Stripe charge object upon success, or a Gift (and a log entry containing '500 Internal') upon failure. But this is irrelevant because the return value is never actually used. Pass by reference, remember? I didn't.

We're now back at Gift#initialize, Rails finally creates a Gift object using the args modified [mostly] in-place by all of the above.

Another step back and we're at GiftSale#create again. This method returns either the shiny new Gift object or an error string (???), and the API logic branches on its type. For further confusion: not all of the method's returns are explicit, and those implicit return values are nested three levels deep. (In Ruby, a method will return the last executed line's return value automatically, allowing e.g. `def add(a,b); a+b; end`)

So, to summarize: GiftSale#create jumps back and forth between Gift five times before finally creating a Gift instance, and each jump further modifies the supplied params in-place.

Also. There are no rescue/catch blocks, meaning any issue with any of the above results in a 500. (A real 500, not a fake 500 like last time. A real 500, with tragic consequences.)

If you're having trouble following the above... yep! That's why it took FOUR FREAKING DAYS! I had no tests, no documentation, no already-built way of testing the API, and no idea what data to send it. especially considering it requires data from Stripe. It also requires an active session token + user data, and I likewise had no login API tests, documentation, logging, no idea how to create a user ... fucking hell, it's a mess.)

Also, and quite confusingly:
There's a class for GiftSale, but there's no table for it.
Gift and GiftSale are completely interchangeable except for their #create methods.

So, why does GiftSale exist?
I have no bloody idea.
All it seems to do is make everything far more complicated than it needs to be.

Anyway. My total commit?
Six lines.
IN FOUR FUCKING DAYS!

AHSKJGHALSKHGLKAHDSGJKASGH.

Going through Master Card API docs to see how to integrate it, saw that they have sample code, checked Java sample code and found this:

String data = MessageFormat.format(
"'{'\"apiOperation\":{0},"
+ "\"sourceOfFunds\":'{'\"type\":{1},\"provided\":'{'\"card\":'{'\"numbe\":{2},"
+ "\"expiry\":'{'\"month\":{3}, \"year\":{4}'}',\"securityCode\":{5}'}}}',"
+ "\"order\":'{'\"reference\":{6}'}',"
+ "\"transaction\":'{'\"amount\":{7},\"currency\":{8},\"reference\":{9},\"targetTransactionId\":{10}'}'," + "\"customer\":'{'\"ipAddress\":{11}'}}'",
apiOperation,
sourceOfFundsType,
cardNumber,
cardExpiryMonth,
cardExpiryYear,
cardSecurityCode,
orderReference,
transactionAmount,
transactionCurrency,
transactionReference,
targetTransactionId,
customerIpAddress );

FOR FUCK SAKE what happened to JSONObject (for Android) class, I'm sure it is a waaaay better solution than that mess ...
And from Oracle:
JsonObject value = Json.createObjectBuilder()
.add("firstName", "John")
.add("lastName", "Smith")
.add("age", 25)
.build();

I guess that is a cleaner understandable solution than what master card has.

Preface: i'm pretty... definitely wasted. rum is amazing.

anyway, I spent today fighting with ActionCable. but as per usu, here's the rant's backstory:

I spent two or three days fighting with ActionCable a few weeks ago. idr how long because I had a 102*f fever at the time, but I managed to write a chat client frontend in React that hooked up to API Guy's copypasta backend. (He literally just copy/pasted it from a chat app tutorial. gg). My code wasn't great, but it did most of what it needed to do. It set up a websocket, had listeners for the various events, connected to the ActionCable server and channel, and wrote out updates to the DOM as they came in. It worked pretty well.

Back to the present!

I spent today trying to get the rest to work, which basically amounted to just fetching historical messages from the server. Turns out that's actually really hard to do, especially when THE FKING OFFICIAL DOCUMENTATION'S EXAMPLES ARE WRONG! Seriously, that crap has scoping and (coffeescript) syntax errors; it doesn't even run. but I didn't know that until the end, because seriously, who posts broken code on official docs? ugh! I spent five hours torturing my code in an effort to get it to work (plus however many more back when I had a fever), only to discover that the examples themselves are broken. No wonder I never got it working!

So, I rooted around for more tutorials or blogs or anything else with functional sample code. Basically every example out there is the same goddamn chat app tutorial with their own commentary. Remember that copy/paste? yeah, that's the one. Still pissed off about that. Also: that tutorial doesn't fetch history, or do anything other than the most basic functionality that I had already written. Totally useless to me.

After quite a bit of searching, the only semi-decent resource I was able to find was a blog from 2015 that's entirely written in Japanese. No, I can't read more than a handful of words, but I've been using it as a reference because its code is seriously more helpful than what's on official Rails docs. -_-

Still never got it to work, though. but after those five futile hours of fighting with the same crap, I sort of gave up and did something else.

zzz.

Anyway.
The moral of the story is that if you publish broken code examples beacuse you didn't even fking bother to test them first, some extremely pissed off and vindictive and fashionable developer will totally waterboard the hell out of you for the cumulative total of her wasted development time because screw you and your goddamn laziness.

THE UNITY API IS SUCH A PILE OF UTTER FUCKING DOGSHIT I CANNOT BELIEVE IT

EVERY FUCKING TUTORIAL IS OUTDATED SINCE LIKE FOUR YEARS

THE FUCKING REFERENCE OFTEN DOESN'T EVEN LIST THE NEEDED ARGUMENTS SO HAVE TO GOOGLE AGAIN

"MOST FRIENDLY ENGINE" MY ASS

GRAAAAAHHHH
NOT TO FORGET THAT ALL EXPLAINING VIDEOS WERE MADE IN 2011 AND ALL VIDEOS ARE PLASTERED WITH ANNOTATIONS SINCE EVERYTHING IS DIFFERENT

Not a rant about anything in particular. Just a summary of some feelings stored in the hateful part of my heart.

Developing for Android: Add this third-party library to your Gradle build. Use (this) built-in Android class to make the thing work.

*Clicks link

Deprecated since API version SUCKMYDICK-7. Use (this) instead

*Clicks link

Deprecated since API version LICKMYBALLS-32. Use...

Developing for Windows: Please use (this) API call. It was literally already available before Bill Gates was born. Carbon dating has placed this item to older than the universe itself and it is likely the entry point for the big bang. It is also still the best way to accomplish (task).

Developing for Linux: "Hmm, I wonder how to use this"

> > > Some shitty mailing list in small blue monospace font tells you to reference a man page that is three versions behind but the only version available.

What? Those three sentences didn't explain it enough? Well, maybe you aren't cut out for this type of thing.

JavaScript: you know how it is.

SQL: You expect a decent-quality answer from stack overflow but you always get an outdated and hacky response and it's using syntax from Microsoft SQL. You need MySQL.

C#: A surprising number of Microsoft forum results ranking high on Google. You click on one in hopes that it will be of any sort of quality. You quickly close the tab and wonder why you ever even had hope.

Literally any REST API: Is it "query" or "q"? "UserID" or "user_id"? Oh, fuck, where's the docs again?

You thought you escaped JavaScript, but it was a trick!: Some bullshit library you downloaded to make your other library work redefined one of the global variables in the project you inherited. Now you get 347 "<x> is not a function" errors in your console. Good luck, asshole.

FontAwesome/ Material fonts/ Any icon font pack: You search "Close" for a close button icon. No results. You search "Simplified railroad crossing sign without the railroad". You get a close icon.

I think that's all of my pent up rage. Each of them were too small for an individual rant so I had to do this essay.

Tutorial video is for sissies. Real men read API Reference

Not just Stack Overflow, but all help forums:

If someone asks a question, and you think you might know the answer based on its subject or context, please spare a thought for others and instead of dragging the post on for pages and pages without an answer in order to show the poster up for being inadequately articulate, please make the first post a link to at least seemingly relevant reference material.

Nothing more frustrating looking for some undocumented api syntax, finding a similar question that you might be able to gleam from, and then only seeing pages of "aaah, but what OS are you on? What colour is your desktop? What is the average velocity of a..."

You can see they are struggling with an api, and you clearly know where to find the answer, so just quit the cocksure jostling of pretending you're any good at support and just be useful. The poster may be a tool, but the thousands of hits from Google will just think you're a twat.

SO MAD. Hands are shaking after dealing with this awful API for too long. I just sent this to a contact at JP Morgan Chase.

-------------------
Hello [X],
1. I'm having absolutely no luck logging in to this account to check the Order Abstraction service settings. I was able to log in once earlier this morning, but ever since I've received this frustratingly vague "We are currently unable to complete your request" error message (attached). I even switched IP's via a VPN, and was able to get as far as entering the below Identification Code until I got the same message. Has this account been blocked? Password incorrect? What's the issue?

2. I've been researching the Order Abstraction API for hours as well, attempting to defuddle this gem of an API call response:

error=1&message=Authentication+failure....processing+stopped

NOWHERE in the documentation (last updated 14 months ago) is there any reference to this^^ error or any sort of standardized error-handling description whatsoever - unless you count the detailed error codes outlined for the Hosted Payment responses, which this Order Abstraction service completely ignores. Finally, the HTTP response status code from the Abstraction API is "200 OK", signaling that everything is fine and dandy, which is incorrect. The error message indicates there should be a 400-level status code response, such as 401 Unauthorized, 403 Forbidden or at least 400 Bad Request.

Frankly, I am extremely frustrated and tired of working with poorly documented, poorly designed and poorly maintained developer services which fail to follow basic methodology standardized decades ago. Error messages should be clear and descriptive, including HTTP status codes and a parseable response - preferably JSON or XML.
-----

This whole piece of garbage is junk. If you're big enough to own a bank, you're big enough to provide useful error messages to the developers kind enough to attempt to work with you.

¡rant|rant

Nice to do some refactoring of the whole data access layer of our core logistics software, let me tell an story.

The project is around 80k lines of code, with a lot of integrations with an ERP system and an sql database.

The ERP system is old, shitty api for it also, only static methods through an wrapper to an c++ library

imagine an order table.

To access an order, you would first need to open the database by calling Api.Open(...file paths) (yes, it's an fucking flat file type database)

Now the database is open, now you would open the orders table with method Api.Table(int tableId) and in return you would get an integer value, the pointer.

Now for the actual order. first you need to search for it by setting the search parameter to the column ID of the order number while checking all calls for some BS error code

Api.SetInt(int pointer, int column, int query Value)
Then call the find method.

Api.Find(int pointer)

Then to top this shitcake of an api of: if it doesn't find your shit it will use the "close enough" method of search.

And now to read a singe string 😑

First you will look in the outdated and incorrect documentation given to you from the devil himself and look for the column ID to find the length of the column.

Then you create a string variable with ALL FUCKING SPACES.

Now you call the Api.GetStr(int pointer, int column, ref string emptyString, int length)

Now you have passed your poor string to the api's demon orgy by reference.

Then some more BS error code checking.

Now you have read an string value 😀

Now keep in mind to repeat these steps for all 300+ columns in the order table.

News from the creators: SQL server? yes, sql is good so everything will be better?

Now imagine the poor developers that got tasked to convert this shitcake to use a MS SQL server, that they did.

Now I can honestly say that I found the best SQL server benchmark tool. This sucker creams out just above ~105K sql statements per second on peak and ~15K per second for 1.5 second to read an order. 1.5 second to read less than 4 fucking kilobytes!

Right at that moment I released that our software would grind to an fucking halt before even thinking about starting it. And that me & myself and I would be tasked to fix it.

4 months later and two weeks until functional beta, here I am. We created our own api with the SQL server 😀

And the outcome of all this...

Fixes bugs older than a year, Forces rewriting part of code base. Forces removal of dirty fixes. allows proper unit and integration testing and even database testing with snapshot feature.

The whole ERP system could be replaced with ~10 lines of code (provided same relational structure) on the application while adding it to our own API library.

Best part is probably the performance improvements 😀. Up to 4500 times faster and 60 times less memory usage also with only managed memory.

The Zen Of Ripping Off Airtable:
(patterned after The Zen Of Python. For all those shamelessly copying airtables basic functionality)

*Columns can be *reordered* for visual priority and ease of use.

* Rows are purely presentational, and mostly for grouping and formatting.

* Data cells are objects in their own right, so they can control their own rendering, and formatting.

* Columns (as objects) are where linkages and other column specific data are stored.

* Rows (as objects) are where row specific data (full-row formatting) are stored.

* Rows are views or references *into* columns which hold references to the actual data cells

* Tables are meant for managing and structuring *small* amounts of data (less than 10k rows) per table.

* Just as you might do "=A1:A5" to reference a cell range in google or excel, you might do "opt(table1:columnN)" in a column header to create a 'type' for the cells in that column.

* An enumeration is a table with a single column, useful for doing the equivalent of airtables options and tags. You will never be able to decide if it should be stored on a specific column, on a specific table for ease of reuse, or separately where it and its brothers will visually clutter your list of tables. Take a shot if you are here.

* Typing or linking a column should be accomplishable first through a command-driven type language, held in column headers and cells as text.

* Take a shot if you somehow ended up creating any of the following: an FSM, a custom regex parser, a new programming language.

* A good structuring system gives us options or tags (multiple select), selections (single select), and many other datatypes and should be first, programmatically available through a simple command-driven language like how commands are done in datacells in excel or google sheets.

* Columns are a means to organize data cells, and set constraints and formatting on an entire range.

* Row height, can be overridden by the settings of a cell. If a cell overrides the row and column render/graphics settings, then it must be drawn last--drawing over the default grid.

* The header of a column is itself a datacell.

* Columns have no order among themselves. Order is purely presentational, and stored on the table itself.

* The last statement is because this allows us to pluck individual columns out of tables for specialized views.

*Very* fast scrolling on large datasets, with row and cell height variability is complicated. Thinking about it makes me want to drink. You should drink too before you embark on implementing it.

* Wherever possible, don't use a database.
If you're thinking about using a database, see the previous koan.

* If you use a database, expect to pick and choose among column-oriented stores, and json, while factoring for platform support, api support, whether you want your front-end users to be forced to install and setup a full database,
and if not, what file-based .so or .dll database engine is out there that also supports video, audio, images, and custom types.

* For each time you ignore one of these nuggets of wisdom, take a shot, question your sanity, quit halfway, and then write another koan about what you learned.

* If you do not have liquor on hand, for each time you would take a shot, spank yourself on the ass. For those who think this is a reward, for each time you would spank yourself on the ass, instead *don't* spank yourself on the ass.

* Take a sip if you *definitely* wildly misused terms from OOP, MVP, and spreadsheets.

I was hired as a wpf dev for a large scale project and I was told that the backend is perfectly complete and that I am able to work right away, then after I started working, I noticed that every request I sent was returning a 500 internal error, I thought it was a problem on my end, but later I found out that they didn't even implement the api yet, it was just a "reference" for devs, which basically means that I should work and give them a perfectly functional app without testing anything at all.

Context: New to typescript. Writing a thing, doing it for work, good opportunity to stretch my dev legs. Using a propriety lib, alternatives not an option.

Rant begin:
SOOOO, who the fuck thought THIS was a good idea:
1. Lib has minified react in dev (because closed source) meaning no downstream errors AND the entire premise of the lib is that a widget is a react component, so I'm writing typescript react the entire time without downstream errors
2. SHIT docs. By that, I mean there's an API reference page that's so sparse there's literally a set of CRUCIAL interfaces that only say the word 'Interface' on them. That's it. that's what i get. It's an interface. NO FUCKING SHIT SHERLOCK, what the fuck is it though? What's its purpose? Is it an interface for a dog? A dog that has a 'shit' property? or a cat? or a cat eating dog shit? Nobody fucking knows - the docs sure as fuck don't care.
3. No syntax highlighting - editors, IDEs (i've tried a few) can't even find the lib inside this environment, so Code and everything else thinks I'm importing shit that doesn't even exist - so no error prediction, code completion based on syntax of the library, none of that.
4. There are some EXTREMELY basic samples - these samples exclusively use React classes - no function components, no hooks, nada - just classes and even perfect replicas of the sample code display erratic behavior like errors about missing props, so that's mostly FUCKING USELESS
5. And this... this is where the straw breaks the fucking camel's back... there's no... there's no hot reloading... Do you know what that (in conjunction with the previous 4 fuckups) means?

When I write anything or I fuck up (which of course I'm doing every time I write half a line because how the fuck?) I have to restart the client and server EVERY FUCKING TIME and manually test to see if the error (THAT ONLY GETS REPORTED IN THE LOCAL UI) is gone or different.
Then, once I see the error, it isn't an error: it's the minified React error-decoder link and guess what? It isn't really clickable a link OR copyable, meaning that every FUCKING time I get a new error, I have to MANUALLY TYPE A FUCKING 50 CHAR URL TO FIND OUT A GENERIC REACT ERROR MESSAGE WITHOUT A LINE NUMBER OR ANY FUCKING CONTEXT. I HAVE TO DO THIS CONSTANTLY TO SEE IF ANYTHING I'M DOING EVEN WORKS.

6. There's no github to complain to the maintainers or search for issues because it's NOT FUCKING OPEN SOURCE so there is literally nothing to be fucking done about it.

This is due in a week and a half, found out about it last Friday. How's your day going?

PS: good to be back after a long respite from dev ranting.

Dev-tip of the day:

Always make sure that your database table fields are long enough to hold the entire third-party API client reference ID, so that the last three characters won’t be cut off..

I’ve been “crying” here while debugging for the last hour because my API function and everything else worked, but the result wasn’t showing up in the third-party application (their API returned a 200), but when I tested it with their API sandbox application, it did show up..

Twilio's API docs suck.

Why can't they just provide a simple API reference will all the endpoints and parameters instead of some hand-holding tutorial with a helper library.

I have a nightmare project that I will probably be ranting about quite a lot in the coming weeks, but I don't want SEO to pick up the specifics on the off chance my peers Google the issues we're facing and my profile comes up.

Let me set up the scene by describing the predicament, and then I'll get to the most outrageous thing I've heard while working at this job. It gives you a CLEAR idea of why we're in this situation in the first place.

Anyways, the nightmare project only runs in IE with compatibility mode set to version 6. So it only runs in IE 6 at the latest.

And it is massive. I'm talking real, real enormous.

The most recent roadblock I ran into while Chrome-ifying it is the extensive use of a browser API that was removed 8 years ago.

It involves synchronous data input and I know for a relatively certain fact there's no way to fix it without combing through every single reference to this API and converting the ones that need sync data (not all of them do) to callbacks. How big of an issue is that?

Well, just one of that 15-ish modules has over 900 references to it. Even just creating a spreadsheet of "commented out / doesn't need a fix / needs fix" for each reference in 1/15th of this project would take days of manual labor.

Here's the rant.

So after discussing this issue in the meeting (we ended on "they don't believe me that we can't just replace it with jQuery") I brought up the next issue. One of our 3rd party libraries is so old it doesn't work anymore and we can't modify that code (it's compiled).

They said that even if it was backwards compatible (no fucking way. This version is like, at least 10 years old, I guarantee it) they can't simply replace it because we don't have a subscription to this product anymore (suggesting we find an alternative).

And I fucking kid you not, this is what happened next.

They then began discussing how this is why you shouldn't use 3rd party code. Because it becomes obsolete and you can't even fix it yourself because it's not yours to edit.

Yes. They said this DIRECTLY after we discussed our 900+ references to a browser API >>REMOVED<< 8 years ago. Yes, they said this about a 3rd party library that receives regular support but is totally FUCKED because we NEVER updated it after adding it and we never even renewed the LICENSE.

What the FUCK

Not the worst, but deserves a mention due to how common it is.

Say your whatever object has a method called Configure. You can infer a lot from the configuration parameters or type that it takes, but for whatever reason something is unclear or doesn't work.

Tooltip from xml comments: Sets the configuration.
Official guide: Sets the configuration <br />.
Technical API reference: Sets the configuration.

I would create a support ticket explaining how this is unclear if I wasn't half expecting the suggested solution to be "you know what I mean".

Payment gateways are such a big pain to implement. Docs say that they will return values A,B,C but what you end up recieving is X,Y,Z.
And don't get me started on the webhooks, man they return values completely different values from the api end points and with no reference what so ever to the fields returned by them.

Wish i could get the documentation writer's address and may be the dev as well!!

So my company has 2 different apps that work using the same API located on 2 different servers.

I mistakenly uploaded both apps to the store with the same server reference.

QA tester alerts me on this issue.

Boss: hey, I the tester said both apps works with server1, ask her more info

Me: (what extra info should I need? I've already been told what's the issue in details, so what's the point?) ok, will do.

Me to QA: hey, do really both apps work on the same server?

QA: yeah

Me to QA: ok, will fix

What the fuck have I accomplished by asking the QA team about it?

Was an internal auditor translating department process to a technical spec for a programmer. We were going to leverage an external company's API which would replace our need to use their slow and buggy web app.

During a meeting, an audit teammate suggested something be changed with the external service we were using. I said we could bring it up with the company but we shouldn't rely on it because we were a small customer even during out busiest month (200 from us vs 10000+ from big banks).

Teammate said we should have our programming team fix it. I made it clear that it was not our side and that to build out the service on our side was beyond our scope. Teammate continued to bring it up during the meeting then went back to desk after meeting and emailed us all marked up screenshots of the feature.

I ignored this and finished writing up the specs, sending them over to the programmer building out the service.

30 minutes later I get a call from programmer's manager who was quite angry at an expanded scope that was impossible (engineers were king at this company. Best not to anger them). Turns out my teammate had emailed his own spec to the programmers full of impossible features that did not reference the API docs.

I feel bad about it now but I yelled at my teammate quite loudly. I said he was spending time on something that was not reasonable or possible and when they continued to talk about their feature I yelled even louder.

Didn't get fired but it definitely tagged me as an asshole until I left. Fair enough :)

During my small tenure as the lead mobile developer for a logistics company I had to manage my stacks between native Android applications in Java and native apps in IOS.

Back then, swift was barely coming into version 3 and as such the transition was not trustworthy enough for me to discard Obj C. So I went with Obj C and kept my knowledge of Swift in the back. It was not difficult since I had always liked Obj C for some reason. The language was what made me click with pointers and understand them well enough to feel more comfortable with C as it was a strict superset from said language. It was enjoyable really and making apps for IOS made me appreciate the ecosystem that much better and realize the level of dedication that the engineering team at Apple used for their compilation protocols. It was my first exposure to ARC(Automatic Reference Counting) as a "form" of garbage collection per se. The tooling in particular was nice, normally with xcode you have a 50/50 chance of it being great or shit. For me it was a mixture of both really, but the number of crashes or unexpected behavior was FAR lesser than what I had in Android back when we still used eclipse and even when we started to use Android Studio.

Developing IOS apps was also what made me see why IOS apps have that distinctive shine and why their phones required less memory(RAM). It was a pleasant experience.

The whole ordeal also left me with a bad taste for Android development. Don't get me wrong, I love my Android phones. But I firmly believe that unless you pay top dollar for an android manufacturer such as Samsung, motorla or lg then you will have lag galore. And man.....everyone that would try to prove me wrong always had to make excuses later on(no, your $200_$300 dllr android device just didn't cut it my dude)

It really sucks sometimes for Android development. I want to know what Google got so wrong that they made the decisions they made in order to make people design other tools such as React Native, Cordova, Ionic, phonegapp, titanium, xamarin(which is shit imo) codename one and many others. With IOS i never considered going for something different than Native since the API just seemed so well designed and far superior to me from an architectural point of view.

Fast forward to 2018(almost 2019) adn Google had talks about flutter for a while and how they make it seem that they are fixing how they want people to design apps.

You see. I firmly believe that tech stacks work in 2 ways:

1 people love a stack so much they start to develop cool ADDITIONS to it(see the awesomeios repo) to expand on the standard libraries

2 people start to FIX a stack because the implementation is broken, lacking in functionality, hard to use by itself: see okhttp, legit all the Square libs, butterknife etc etc etc and etc

From this I can conclude 2 things: people love developing for IOS because the ecosystem is nice and dev friendly, and people like to develop for Android in spite of how Google manages their API. Seriously Android is a great OS and having apps that work awesomely in spite of how hard it is to create applications for said platform just shows a level of love and dedication that is unmatched.

This is why I find it hard, and even mean to call out on one product over the other. Despite the morals behind the 2 leading companies inferred from my post, the develpers are what makes the situation better or worse.

So just fuck it and develop and use for what you want.

Honorific mention to PHP and the php developer community which is a mixture of fixing and adding in spite of the ammount of hatred that such coolness gets from a lot of peeps :P

Oh and I got a couple of mobile contracts in the way, this is why I made this post.

And I still hate developing for Android even though I love Java.

Sydochen has posted a rant where he is nt really sure why people hate Java, and I decided to publicly post my explanation of this phenomenon, please, from my point of view.
So there is this quite large domain, on which one or two academical studies are built, such as business informatics and applied system engineering which I find extremely interesting and fun, that is called, ironically, SAD. And then there are videos on youtube, by programmers who just can't settle the fuck down. Those videos I am talking about are rants about OOP in general, which, as we all know, is a huge part of studies in the aforementioned domain. What these people are even talking about?
Absolutely obvious, there is no sense in making a software in a linear pattern. Since Bikelsoft has conveniently patched consumers up with GUI based software, the core concept of which is EDP (event driven programming or alternatively, at least OS events queue-ing), the completely functional, linear approach in such environment does not make much sense in terms of the maintainability of the software. Uhm, raise your hand if you ever tried to linearly build a complex GUI system in a single function call on GTK, which does allow you to disregard any responsibility separation pattern of SAD, such as long loved MVC...
Additionally, OOP is mandatory in business because it does allow us to mount abstraction levels and encapsulate actual dataflow behind them, which, of course, lowers the costs of the development.
What happy programmers are talking about usually is the complexity of the task of doing the OOP right in the sense of an overflow of straight composition classes (that do nothing but forward data from lower to upper abstraction levels and vice versa) and the situation of responsibility chain break (this is when a class from lower level directly!! notifies a class of a higher level about something ignoring the fact that there is a chain of other classes between them). And that's it. These guys also do vouch for functional programming, and it's a completely different argument, and there is no reason not to do it in algorithmical, implementational part of the project, of course, but yeah...
So where does Java kick in you think?
Well, guess what language popularized programming in general and OOP in particular. Java is doing a lot of things in a modern way. Of course, if it's 1995 outside *lenny face*. Yeah, fuck AOT, fuck memory management responsibility, all to the maximum towards solving the real applicative tasks.
Have you ever tried to learn to apply Text Watchers in Android with Java? Then you know about inline overloading and inline abstract class implementation. This is not right. This reduces readability and reusability.
Have you ever used Volley on Android? Newbies to Android programming surely should have. Quite verbose boilerplate in google docs, huh?
Have you seen intents? The Android API is, little said, messy with all the support libs and Context class ancestors. Remember how many times the language has helped you to properly orient in all of this hierarchy, when overloading method declaration requires you to use 2 lines instead of 1. Too verbose, too hesitant, distracting - that's what the lang and the api is. Fucking toString() is hilarious. Reference comparison is unintuitive. Obviously poor practices are not banned. Ancient tools. Import hell. Slow evolution.
C# has ripped Java off like an utter cunt, yet it's a piece of cake to maintain a solid patternization and structure, and keep your code clean and readable. Yet, Cs6 already was okay featuring optionally nullable fields and safe optional dereferencing, while we get finally get lambda expressions in J8, in 20-fucking-14.
Java did good back then, but when we joke about dumb indian developers, they are coding it in Java. So yeah.
To sum up, it's easy to make code unreadable with Java, and Java is a tool with which developers usually disregard the patterns of SAD.

When you're using openapi generators and stuff for generating SDK code and let "the architect" handle the data structure and nomenclature, don't you hate having to add 33 (I counted) models, most of which are just the same class with different name or one property apart from each other, serialization of which gives request body overhead 56-132x (actual calculated results depending on the model complexity) the size of actual data you want to send, just to add support for one endpoint that needs just one model that started this whole madness?
I just had to add this one top level model reference and this happened to me. Those 33 models are not including the ones I already had included in my project so they didn't have to import them again.
For the love of <your_belief_here /> and all that's holy, never ever agree on generating code based on openapi if the person responsible for that is unexperienced. It will do more harm than good, trust me.
Before we decided to go with generated SDK my compiled product was a bit over 30KB, and worked just fine, but required a bit of work on each breaking API change. Every change in the API requires now 75% of that work and the compiled package is now over 8MB (750KB of which is probably my code and actually needed dependencies).
Adding an endpoint handler before? Add url, set method and construct the body with the bare minimum accepted by the server
Now? Add 33 models (or more), run full-project find&replace and hope it will work with the method supplied by the generated code, because it's not a mature tech and it's not always guaranteed it will work.

What happened to the Vue.js docs? Years ago the v2 docs were amazing, simple yet clear and provided detailed information for every API. I picked up v3 after a few years of only using React and the docs are a mess. Guides are not in a logical order, and the API reference is missing a lot of information and forces you to dig through type declarations in the code to figure things out.

On the office we have began to discuss which is best practice in a REST API when reference other objects.

Things like:

A)

```
{
"id": 1,
"field1": "value1",
"referenced": <id>
}
```

or

B)

```
{
"id": 1,
"field1": "value1",
"referenced": {
"id": <id>
}
}
```

I prefer B. What do you think?

a) No documentation. At least you know from the start where you are.

b) Light documentation. At least you might have an API reference or something

c) Badly translated but complex documentation. Had this when I worked for a car manufacturer. Docs were badly translated and actually gave you a dangerously wrong or opposite description of what things really did. It was mental!

Utility libraries, because I actually get to see my life improve because I have them. Creating new projects becomes easier because I put parts I reuse somewhere else.

There was an old config file generator/manager I kept using for a while, some string conversion libraries between formats, some REST/WebRTC API wrappers, I have a web audio API I create tunes with in various projects. CI/CD scripts for laziness so I never have to know how to set anything up again. Lately the thing I'm most happy about is I turned some free text saving service into a makeshift database and it's been working well for about half a dozen projects now. Wouldn't handle large amount of users but can't beat free and easy.

I also find merit in prototypes/old projects, because I can reference random things I did in them in newer unrelated projects. Things too small to warrant their own utility library, argh!

why google???? I mean how hard for them to update a reference document. The reference of gmail api for python2 and python2 is in EOL. The first error I found for python3 is reported in 2015 on GitHub and now is 2019. I mean, seriously the bigB_Company don't have time to update their document. I scratched my head for an hour, what the fuck did goes wrong.

I was given a project to fix and improve a legacy unity VR project I was told was for the oculus rift Now the problems started almost immediately partly stemming from the fact I’ve never used unity before this project was handed to me as my long term TA assignment
And partly from the fact there was no oculus integration in the game at all. it was built for GoogleVr and most of the code the last person wrote consisted of massive sections (25-50 lines) of commented out code and no explanation of what the hell the non-commented parts are supposed to be doing

So long story short. I’m now in a basic unity course, six feet deep in documentation trying to read resources that go way over my head in understanding, and am rebuilding the project from basically scratch (took the assets and saved the c# scripts for reference) and have finally figured out how to at least get the player character constantly moving forward and stream in the WRLD3D environment like the last guy did. Now to get the player character to turn and change direction when the player turns their head with the oculus headset

By the way. WRLD3D is a really cool api thing in my opinion

We were refining a tech debt issue about aligning the names and types of the same reference id on different response models. This is to not confuse our API users and make it more intuitive.

Discussion was wrapping up as we all agreed it was a no-brainer and pretty straight forward.

Then suddenly, one colleague goes: "But what's the benefit?"

Errrm...