Anyone else hate not being a programmer?

A good machine learning algo is what you're looking for, and even then it wouldn't be 100%. For example, think WordAi's NLP algo, but obviously you don't need anything close to that advanced. I taught myself some ML a couple years back, but have since forgotten everything.

Failing that, you're basically stuck creating an index of keywords, then matching each entry up against the # of words and matches it has to other entries. So take a string, strip out the useless words (a, and, the, free, plus, etc.), and generate an index of all words. Then compare each string. If one string has 5 words, and matches 4 of those words in another string, good chance it's the same product.

Again though, machine learning is the better way, but definitely more difficult way.
 


rusvik said:
No excel, all has to be 100% automatic, or at most something like a backend to approve/change those in question.

It's mostly api's yes, what is the last sentence about? %xxx% limit etc?
Click to expand...


How are you looking to manipulate/display the data? If you can store all of the data with the same fields populated, then you can display the data you want by querying the database.

So say you wanted to find Nike Air Jordans, you could do something like

select * from scrapedtable where brand like %nike% and product like %air jordan%

What this would do is return the results where the column brand has a string that contains nike, and the column product contains air jordan. % is a sql wildcard for any values. Limit just says show the first 50 results.
 
Regexp would really empower you.

Just looking at github for a sec I found this node.js javascript library that illustrates a couple concepts with regular expressions.

https://github.com/VerbalExpressions/JSVerbalExpressions

This is for validating a url, but notice the sequence of events.
-go to start of line.
-look for "http". This could be brand keyword
-look for "s" but ok if not found. This could be brand keyword variation
-etc...

But you can create your own set of criteria.....

Code: 
// Create an example of how to test for correctly formed URLs
var tester = VerEx()
            .startOfLine()
            .then( "http" )
            .maybe( "s" )
            .then( "://" )
            .maybe( "www." )
            .anythingBut( " " )
            .endOfLine();

// Create an example URL
var testMe = "https://www.google.com";

// Use RegExp object's native test() function
if( tester.test( testMe ) ) alert( "We have a correct URL "); // This output will fire
else alert( "The URL is incorrect" );

Code: 
// Create a test string
var replaceMe = "Replace bird with a duck";

// Create an expression that seeks for word "bird"
var expression = VerEx().find( "bird" );

// Execute the expression like a normal RegExp object
var result = expression.replace( replaceMe, "duck" );
alert( result ); // Outputs "Replace duck with a duck"
 
there has to be something identical that each product with varying names share. maybe they're all pulling the same product images from the supplier, you could do a MD5 hash on them. or SKUs as vgeek said. exhaust all these options before you try matching on the product titles.
 
As everyone else has said, trying to match the titles is going to be a pita and require extra time. No reason you can't match on something universal like an SKU.
 
This thread is too good to just let die. I have come to a bit of a crossroads and your input would be appreciated.

So I have dabbled around with code for a few years. I can read and get the gist out of most of the code I run across. But I am not good enough to write much of a program from scratch. Have fucked around writing scrapers in python (using scrapy) and hacked up my share of wordpress themes.

My latest obsession is learning how to write iOS apps. So I have been digging into swift courses on treehouse. Which leads to my real problem. I figured swift is the future for iOS development so I figure a guy should start there. It's fairly likely it will become an open source project and be useable outside the apple ecosystem. But who knows.

And this is where my problem lies. Swift is basically a lazy mans hacked up newb version of C. Which leads me to think if I really get into swift, I won't be able to go back to actually learn C. Objective-C has the same problem and will die off in the next 5 to 10 years.

I have actual salable ideas for enterprise iOS apps today. Will I make much on them? Probably not. Is that short term gratification likely to fuck up learning a truly high level language? Do I put my iOS hopes and dreams on hold for 6 months or a year and really crack down and wrap my mind around C? Or just say fuck it and just build my apps? This probably seems dumb, but I really don't know what to do next.
 
shindig said:
Someone else will have a better answer, but is there any reason you want to limit yourself to iOS?

If you just want to make apps, and actually complete them, there's no shame in stuff like xamarin Mobile Application Development to Build Apps in C# - Xamarin

Or if you want to make games, use a game engine. It's hard enough to get art, design, content, features, etc all polished regardless of hardware.
Click to expand...

That is the crux of it. I don't want to be limited by ios I feel like I need to ditch it for now, and stop fucking around with all this other shit. Really put the nose to the grindstone and really dig into C. I can already tell that this swift is going to fuck with my ability to learn a proper language after seeing all the shortcuts Apple is taking. I don't know that shortcut is te right word. But it's a serious issue either way.
 
How good of a developer to you want to be? If you want to be good, then suck it up, and start putting in the hours to learn C, and learn it properly as well. You'll know you're a good developer once you realize the language you're typing code in doesn't matter, and when you can pick up a new language in a matter of a week or two.

Get the fundamentals down, as it's almost close to all the same, regardless of language. The good thing about C is you're going to learn about strict data types, and just that should be enough to make you do it.
 
Programming has to be hands down the best mental exercise out there. There was a period where I stopped programming for a year or two, and I noticed all my other cognitive abilities such as memory declined. I was still running a business, but business is vague, there aren't clearly defined problems and challenges for your brain. Programming is constant problem solving, it's making use of brain plasticity at it's finest.

Since getting back into it, I feel like my mind is at the peak of it's game.

Fuck that "I outsource programming because time is money" bullshit, you probably sit around jerking off most of the time anyway pretending you're Steve Jobs. Programming will increase your overall mental capacity, and keep your brain healthy into old age.
 
Kiopa_Matt said:
Pro tip: If you're going into software, you have a few choices:

a) have solid business acumen + vision as well, and be able to setup and maintain your own shop.

b) get a shitty 9-5

c) specialize in a specific niche. 15 years ago, being a software developer was great, and paid big. $40k/month profit was no big deal back then. Nowadays, globalization has taken affect, and it's totally different. You're not going to make sweet fuck all being an all-purpose, general software developer. You're most likely going to be struggling to land shitty $40/hour contracts, which is a fucken joke considering the amount of training and skill required to produce quality software.

So if you're doing software and are an independent, find a speciality, and dominate it. Instead of just being a software developer, I'm now a developer who specializes in bitcoin and virtual currency. My income has tripled if not more, and by no means am I worried about shortage of work. I equate it to the same as being a doctor. You can be a general practioner at a local hospital, or someone who specializes in a specific field (feet, heart, skin, penis, vagina, whatever...). Guess who gets paid more?
Click to expand...

would you consider asp.net mvc(or ruby/rails or python/django) a for building web apps a specialization that would pay well with contracts?

Kiopa_Matt said:
How good of a developer to you want to be? If you want to be good, then suck it up, and start putting in the hours to learn C, and learn it properly as well. You'll know you're a good developer once you realize the language you're typing code in doesn't matter, and when you can pick up a new language in a matter of a week or two.

Get the fundamentals down, as it's almost close to all the same, regardless of language. The good thing about C is you're going to learn about strict data types, and just that should be enough to make you do it.
Click to expand...

is "c" the only option. Cant you learn programming well with a higher level language like c# or java? curious.

I know they would abstract the memory allocation, would be managed and keep you focussed on problem solving more than machine details.

But I would like to know why "c" or go deeper into assembly when your goal for learning programming is to make complex and scalable web/desktop/mobile apps which would play with data, data bases, output to the UI etc.

dont u feel learning c#/asp.net or ruby/rails or python/django or java/ java web framework in depth would make u an excellent programmer and an expert web application developer(with practice)?
 
John__N said:
would you consider asp.net mvc(or ruby/rails or python/django) a for building web apps a specialization that would pay well with contracts?
Click to expand...

No, being able to build web apps (the language doesn't matter) is not specialization.

I can go on oDesk right now and find 400 people in Bangladesh willing to work for $2 and hour that will build me a web app.

Will it work? Yer.
Will it be optimized? Nope.
Will it scale? Bahahahahahaha.

The only thing that counts now with web apps is speed.

Let me repeat that, the only thing that counts is speed!

How many requests can I serve and how fast can I do it.

If you want to specialize in building web apps, focus on how well you can optimize the app.

Anyone can build a web app, but very few take the time to learn how to optimize them.

John__N said:
But I would like to know why "c" or go deeper into assembly when your goal for learning programming is to make complex and scalable web/desktop/mobile apps which would play with data, data bases, output to the UI etc.
Click to expand...

There are edge cases where knowing C or ASM is an advantage when building web apps.

For example, Real Time Bidding (RTB) when the server has to respond in less than 100ms.

But generally the speed benefits of a web app written entirely in C or ASM are outweighed by time it takes to write them.
 
Just my personal opinion (which is like an asshole, we all have one).

ASP.net -- I would run like hell from it. I didn't even now people still use it. If you're going in web devel though, probably one of the worst choices for a language.

Yeah, C# / Java should be fine, as well as C/C++. The reason I said learn C is due to the strictness of the language. The compiler will be constantly throwing error after error at him, driving him insane, and making him want to throw his laptop at a cement wall. This is a good thing though, because it teaches you how to type code properly.

On the flip side, although I use PHP for majority of my dev work, it's probably THE single worst language to learn from. Simply because PHP lets you do whatever the hell you want, and will let you get away with it. You can turn an integer into a string, float, or even array, and PHP won't complain. PHP is a great language, but you should learn from something better first. At the same time, wouldn't really recommend Python for learning either, as it lets you get away with quite a bit too. Not as much as PHP, but still a fair amount.

I'll unfortunately have to totally disagree with acidie above. Speed doesn't matter so much. Neither does your ability to type code. What matters is your creativity, problem solving ability, and your capability to turn client's visions into realistic, usable online apps. Anyone can learn to type code, but being able to take say a 12 page set of specs, and turn it into a professional app that can be used across an organization with ease is what you're going to get paid for.

There's a huge difference between someone who can code up a little WP plugin, or hack together a Python script to do this or that, versus someone who can organize and put together an enterprise level system, while also helping consult and contribute ideas to process / user flow, etc. If you want high paying gigs, you're going to need to be capable of the latter.
 
Kiopa_Matt said:
ASP.net -- I would run like hell from it. I didn't even now people still use it. If you're going in web devel though, probably one of the worst choices for a language.
Click to expand...

Are you talking about ASP.NET Forms or ASP.NET MVC or ASP.NET WebAPI?

Kiopa_Matt said:
The reason I said learn C is due to the strictness of the language. The compiler will be constantly throwing error after error at him, driving him insane, and making him want to throw his laptop at a cement wall. This is a good thing though, because it teaches you how to type code properly.
Click to expand...

You know C# and Java are strongly typed right?

Kiopa_Matt said:
Simply because PHP lets you do whatever the hell you want, and will let you get away with it.
Click to expand...

There are a million reasons never to use PHP for anything, but I concede this is one main ones.

The other is lack of threads... queue a million Indian's tears of pain.

Kiopa_Matt said:
Speed doesn't matter so much.
Click to expand...

Tell that to a user waiting 12 minutes for that SQL query to return.

Kiopa_Matt said:
while also helping consult and contribute ideas to process / user flow, etc.
Click to expand...

Because you'll need to be a fucking expert in deflecting blame, I mean "consult and contribute ideas" when the PM asks why the app crashes after running a single query.
 
Kiopa_Matt said:
Just my personal opinion (which is like an asshole, we all have one).

ASP.net -- I would run like hell from it. I didn't even now people still use it. If you're going in web devel though, probably one of the worst choices for a language.

Yeah, C# / Java should be fine, as well as C/C++. The reason I said learn C is due to the strictness of the language. The compiler will be constantly throwing error after error at him, driving him insane, and making him want to throw his laptop at a cement wall. This is a good thing though, because it teaches you how to type code properly.

On the flip side, although I use PHP for majority of my dev work, it's probably THE single worst language to learn from. Simply because PHP lets you do whatever the hell you want, and will let you get away with it. You can turn an integer into a string, float, or even array, and PHP won't complain. PHP is a great language, but you should learn from something better first. At the same time, wouldn't really recommend Python for learning either, as it lets you get away with quite a bit too. Not as much as PHP, but still a fair amount.

I'll unfortunately have to totally disagree with acidie above. Speed doesn't matter so much. Neither does your ability to type code. What matters is your creativity, problem solving ability, and your capability to turn client's visions into realistic, usable online apps. Anyone can learn to type code, but being able to take say a 12 page set of specs, and turn it into a professional app that can be used across an organization with ease is what you're going to get paid for.

There's a huge difference between someone who can code up a little WP plugin, or hack together a Python script to do this or that, versus someone who can organize and put together an enterprise level system, while also helping consult and contribute ideas to process / user flow, etc. If you want high paying gigs, you're going to need to be capable of the latter.
Click to expand...

i know nothing about coding, but when i read that line, i knew a train wreck was in progress.
 
You must log in or register to reply here.
Top Bottom