Topic: Closed (Read 8672 times)

Your CEO of a company do you use .NET or have your purchased a Windows server OS to run it? If yes then you know they require you to purchase a SQL server software license. So obviously you have no idea.

Why are you recommending a LIBRARY for java. Bitcoinj is just a library to talk to the bitcoin network and have a wallet. You would have implement JSON-RPC or even a Rest API to get information and send coins. I would NOT recommend in any WEB based bitcoin project. They should use Bitcoind. While I know you could write this into a java web application I would also recommend that isn't good, a bug on the website can easily open it up to being attacked and robbed.

This thread honestly should put in the trash too much false information don't listen to anything in here and do your own research.

Don't blame me cause you got owned by DeathAndTaxes and are now changing your view. ASP.NET Framework unless you have a lot of time to make it work like pof which to be honest is not worth it and very lazy will never scale like php and python. To be honest (Probably going to start a flame war) I hate python, I hate the syntax, and I would pick python over ASP.NET for any web project today.

That is 100% false and no matter how many times you say it, it doesn't get any truer.

A SQL Server license is 100% NOT required to use .NET.  Period.  One can optionally license SQL Server and no license is required at all to develop on the .Net Framework.  Many .Net applications use MySQL as a datastore.  To use SQL Server requires a license.  To use SQL Server and not use .Net still requires a license.  Not sure how many times people have to say it before you get it.

.Net Framework =/=  SQL Server

Your ignorance is showing.  

What are you talking about? I thought this was dead but then you feel the need to throw in new falsehoods. .NET is a software framework.  Nothing more.  

It can connect to just about any db on the planet.  No need for custom libraries.  Out of the box, day 0, in visual studio you can build a new MVC project and setup MySQL as the datastore in all of about five minutes for example.  ADO.NET, LINQ, and Entity Framework all work fine with MySQL or even Oracle (if one felt so inclined).  Data access isn't slower for non-Microsoft RDBMS.

gweedo, I read your other posts. That one in particular had wrong information though. Regardless if you made factually correct statements in the other posts, that one was just so wrong that I really wanted to reply and make corrections, despite it being old.

Which points specifically do you want me to prove? Any developer who understands scaling a database driven app would understand the concept in #1. It's called scaling up vs scaling out. I'm not going to dig up links for you, feel free to google it yourself.

#2 is also a fact. .NET is not a database, it is a software platform. You can access any database from .NET apps, including MongoDB, Cassandra, MySQL, MariaDB, PostGresQL, etc. Again, feel free to find links that disprove this. Just search for any of those db names + "C# driver" or "C# client" and you'll find the drivers for connecting to those db's.

#3 is also well understood in the programming community. It's well known that statically typed, compiled languages tend to be faster than dynamic ones. That doesn't make them better than dynamic ones. In fact, people often argue that the increased productivity and flexibility of using dynamic languages like python and ruby outweigh the benefits of typed/compiled ones. See any discussion about javascript  and typescript, C# vs Ruby, C# vs PHP.

I'm presenting some alternate viewpoints for you here. It's up to you if you want to stick with the uninformed ones you are holding or if you want to look into what I'm trying to clue you in on. I'm definitely not saying .NET is better than any other language. I think those types of statements are stupid no matter what language they are for or against.

One valid answer is that a developer might be very proficient in a particular language. It takes only a couple of weeks for a decent developer to start putting out code in anew language, but the quality of that code isn't going to be very good, and the speed of development will be much lower than if they used their "native" language. It takes about 6-8 months for a developer to become proficient in a new language. So if you are a developer and you are already proficient in one technology, that's one big +1 to the PRO/CON list of which tech to build a new project with. There are other things to consider obviously, but it's a pretty siginificant factor. I can build .NET and python projects VERY fast. It would take me at least 4x longer to build the same thing, at the same level of quality in PHP. That doesn't mean PHP is a bad choice, nor does it mean .NET or Python are the best choices for everyone. In my case, Python or .NET would probably be the best choices.

Other things to consider are licensing costs, but if you are a small startup that's not that significant. If you build for the cloud, the cost of a monthly instance of a windows web server vs a linux one is almost the same. If you use BizSpark you get to keep 2 sql server and 4 windows server licenses in production even after you graduate from the program. My approach when building on .NET is to use pgSQL when possible, or just stick with SQL server (SQL Azure is cheap, monthly fee), and build it in such a way that I could change the database to pgSQL or something else if I need to migrate to something to avoid the SQL server fees.

I am a professional developer. I currently work in Python, but have worked in Ruby, Java, and also have years of experience with .NET. I just wanted to let anyone who is reading this that gweedo flat out doesn't know what he is talking about.

1. Stackoverflow chose to have a scale up architecture. I am skeptical that this would be a good choice for most businesses, and am waiting to see if eventually their gamble will fail. They essentially have bet that their traffic will not grow faster than hardware can. So far that has proved to be true. This has nothing to do with choosing .NET, and more to do with choosing a scale up approach for their database.

They would have the same exact problem if they tried with MySQL, and MSSQL is more proven when it comes to high performance at scales that large (even when you "just swap out some software" as you put it, which by that I assume you mean optimize which storage engine you're using, but that alone still won't let MySQL scale UP better than pgSQL or MSSQL).

2. This has nothing to do with their choice of using .NET. They are using SQL Server as their database. Yes, .NET tends to go hand in hand with SQL server but there is no reason that has to be so. There are lots of projects that use .NET with an open source database like pgSQL or MySQL.

3. .NET is a compiled language, and statically typed. The compilers are very sophisticated, and the language features are very advanced. .NET actually tends to scale and perform better than dynamic languages like Ruby, PHP, and Python.


The scary part about the internet is it lets people who have no idea what they are talking about propagate bad information. I'm sure gweedo means well but his posts in this thread as uninformed. This is how .NET gets such a bad rap. I think all languages are suitable for mos tasks. I stay away from Ruby because it's become the language where all beginners flock to, and they seem to have a lot of security flaws lately. Python and .NET are my two goto languages now. As long as you have developers who know what they are doing, they can make any project work and scale in their language of choice. So most language vs language arguments are stupid to begin with, but they are still full of people spouting off incorrect information.

"Rapid application development" is precisely what got me 15+ years ago. Redmond does a fantastic job of making it EASY to develop (there is NO question about that). I don't have any experience with C#, but I did use Visual C++ for many years and it was an absolute pleasure. At the time, there was nothing else like it, so there wasn't really a decision to be made.

But I call it a TRAP. A licensing trap, that once you commit to is brutal to get out of.  I vividly remember my transition off of Redmond's tech and it involved an exorbitant amount of caffeine and extra strength visine coding the nights away. When I came out the other side, it was absolute PARADISE.

Today, the decision in my opinion, has swayed the other way. Open source toolkits are very mature and extremely well supported. They ARE still lacking as in the case of Mono vs. .NET, but you make due given the alternative. I think the real question is, "WHY would you EVER want to use closed source technology?" There are exceptions, which I acknowledge, but they are very, VERY far and few between.

Well you combined a bunch of things as if they are a single entity.

C# is an open international standard.
http://www.ecma-international.org/publications/standards/Ecma-334.htm

Mono is an open implementation of .NET.
http://www.mono-project.com/Main_Page

MS SQL Server is closed source however to answer the question "why"?  Rapid application development.  Nice tools for rapidly constructing entity framework when using Visual Studio and MS SQL Server.  Of course it is somewhat of a strawman argument as it isn't required.  .Net will work fine with just about any datstore SQL or otherwise.

I would like for someone to make a sane argument as to why a startup should use proprietary closed source software like .NET ASP/C# and MSSQL.

Problems was maybe a poor choice of words.  It has its "disadvantages".  I use several StackExchange sites on a daily basis (particularly StackOverflow and Android).  They are truly great for getting "quick" answers to your questions (usually without even asking, because they've already been answered). But its a Q&A format, there is little to no discussion (merely comments).

When I used the word problems, it was really in reference to how heavily moderated those sites are.  It makes for a much cleaner site (which is nice), but I truly hate having every question I write being adjusted by someone who thinks it should be worded THEIR way (I understand why they do it, but I still find it incredibly annoying). I for one prefer the RAW freedom of expression available from forums like BT (for better or worse). In the end, they ALL serve their purpose.

What problems does bitcoin stackexchange have?

I don't think I expressly recommended bitcoinj over bitcoind. I simply threw it in to cover all "reasonable" options if you wanted to implement your own blockchain services.


These FORUMS are filled with trash (and more trolling than I care to stomach). People expressing their own opinions, for better or worse, and sifting thru the ruble seems to be the only way to get up-to-date information regarding Bitcoin. If you're aware of better source than bitcointalk, I'd sure like to know as well. Bitcoin StackExchange has its own problems.

Lol, just joking around... I know I am mostly to blame for derailing this thread...

Yep, there's a good reason why good programmers are so expensive

By now, what should be abundantly clear is that: to each is own.  In regards to the front-end, I will actually concede that even ASP.NET is suitable if that is your strength (and you are aware of the licensing requirements, if you decide to extend Redmond's tech into your back-end). However, personally, I would go with either PHP (which is insanely popular in the Bitcoin community and extremely well supported) run over Nginx OR Node.js if you're comfortable using JavaScript and would like to assure maximum scalability on modest hardware.

Keep in mind that with ANY Bitcoin-related business, you will eventually have to tie back into the blockchain at some point.  There are companies like BitPay that do a very good job of abstracting a simple payment gateway over the blockchain and is a good solution if you're only looking to accept payments. However, if you need to manipulate the blockchain in anyway (like when sending coins) or integrate with it (like when you need to track transactions and confirmations) then blockchain.info provides a great API that lets you do this with little fuss.

If you decide to run your own back-end daemon (maybe for security reasons), you are certain to receive a lot of mixed advice as to the pros and cons of virtually every language, framework, platform, OS in existence (at least from the people on this thread). Stay away from Windows (at all costs) and go with whatever Linux distro you feel comformatable with (Ubuntu and CentOS are my favorites). There is an official C-based Bitcoin daemon as well as a very popular Java alternative. I'd stick with these for now as the others are very much unfield-tested at the moment.

Good luck!

Given your choice of languages and frameworks, which would you use in building a Bitcoin-related web application at this moment?