Using a Microsoft language has given us huge productivity benefits (most notably Visual Studio and Azure support). There is a fantastic F# community (http://fsharp.org) which is untethered to Microsoft, and it's exciting to see the language grow as an open-source and cross-platform tool.
Visual Studio seems to be the only real benefit I'm seeing (and I agree, it's a very major benefit).
In my view the .net libaries are probably the biggest advantage over OCaml and Haskell. Similar to how access to Java libraries gives Clojure an advantage over lisps like Common Lisp and Racket.
Oh, and if you want to use Windows without mingw then F# is a far better choice than OCaml or Haskell. The OCaml packaged manager doesn't even work properly on Windows, as far as I'm aware.
Might be worth checking out bizspark , as I know it lists lots of startups that are using a .net stack.
Also probably worth mentioning what you're compiling the list for if you want more responses from HN users...
More info: We actually don't host on Azure (Web Sites not flexible enough for us), but use a mixture of AppHarbor  (awesome product and team), and various Amazon web services.
 - https://www.openrent.co.uk
 - https://appharbor.com
Would you care to share any specifics? I'm curious to know what you're trying to do that doesn't work on Azure.
One example - if you have any processing in the application_startup, Web Sites won't work for you. Applications are randomly killed and restarted, and don't allow for a startup time, meaning outages when this happens. Customer support suggested using a fail over to work around this, but the fail over only kicks in once your app has been down for a certain amount of time - so you'll still have down time.
Sure - we could re-architecture the app so that isn't a requirement, but then I'm making design decisions based on Azure constraints. Why would I when there are alternatives like AppHarbor?
Azure is a bit hit and miss. Its brilliant for getting something up an running quickly (using websites / SQL Server), but is a little flaky at scale.
Key problems include connection issues with SQL Server, connection issues with their hosted Redis service, pricing of SQL Server when using advanced features like geo-replication.
All in all though, its a pretty good development experience once you get your head around the fact that in the cloud services fail and there is nothing you can do about it except plan for it.
Oh and the Bizspark program they have gives you $100 worth of free hosting on Azure which is always nice.
What sort? If it's intermediate connection problems .NET 4.5.1 added Connection Resiliency to ADO.NET . If you're using a recent enough version of Entity Framework it goes even further .
public class MyConfiguration : DbConfiguration
() => new SqlAzureExecutionStrategy(1, TimeSpan.FromSeconds(30)));
We have our own retry logic, which also logs the issue so we are aware of how frequently errors occur while a command / transaction is being executed.
This is using SQL Azure with the "Business" tier, so it will be interesting to see how the new (much more highly priced tiers) Standard and Premium tiers go.
There are a lot more .NET start-ups than one is lead to believe, but you don't hear so much about them because they're too busy making money to give interviews to TechCrunch ;) Seriously though, they tend to be from developers branching out of existing enterprises & "non-cool" fields and have existing customers or bootstrap funding.
With Azure, AWS, Bizspark & MAPSDD cost implications of being on the .NET stack are negligible comparing to the long term savings in maintainability.
Since it was asked what happens if the BizSpark period is over, let me state on detail about the model:
In BizSpark, its true that you go vendor lock in as long as you are building strict dependency on their platform, lets say if you use EF for accessing SQL, you may have a problem in the long run.
Even if you've left BizSpark, a MS Server License is to be paid once? (not per year, not every two years etc.)
I do not understand where the myth comes from that one has to pay every year or continuously to use MS licenses?
The math is simple:
A W2K12 Standard server license costs around 850 USD (~ 700 EUR) - once - even if Opensource would is cheaper in term of license costs, a software running 24/7 for a one-time payment of 700 EUR isn't that expensive.
Now, it comes down to scale up or scale out:
- in the one scenario, because of no costs you can scale massivley out, going from one machine to n machines
- but if you ever tried to administrate your machine park lonely(!!) if it counts 5+ machines (while you have to develop/maintain & run the app itself, too!), some may consider to hire a network admin (which is a fixed salary per year which usually outweighs the costs of your MS Server license)
The reason why C# is used, is simple (and was stated above):
- C# and .NET are very convenient and mature environments to write apps in
- C# is considered by me as a beautiful language, i really like it (coming from x years of "C-styled-syntax-languages" could make me somehwat biased ;-)
Instead of the licensing costs of MS products, i see much more important difficulties in the mid- to long-run future if someone built on it:
MS maybe not survive the upcoming decade because of huge market shifts; then, in 10 years, it may be a problem if you will not get any support because the company maybe no more existing.
God thanks, the next version of ASP MVC has a VM for *nix-derivatives, so you could solve this issue by running your app on a Linux-box.
(And if you then have built your app additionally on a top of an ORM, you won't have any issues)
Microsoft has a 3-year release cycle for their products. You'll end up paying every three years. For every single product. Microsoft's biggest cash cows are it's OS and Office products and always have been.
> A W2K12 Standard server license costs around 850 USD (~ 700 EUR)
Assuming you have just one server. At the least you will need a production server and a development server. And at least one for every developer. That's one OS license, one SQL Server license and one Visual Studio license per developer. Personally, I love work with a lot of virtual machines off Vagrant for development, testing and deployment. I am not sure how that works with MS licences.
> if it counts 5+ machines [...] some may consider to hire a network admin
It's fairly easy to manage ~100 machines using Ansible, Docker, Chef, etc. But in the Microsoft stack you probably want to keep the number of machines much lower.
> scale up or scale out
Scaling up is the right approach on Windows platforms but here is the issue I have with it. You cannot add RAM on the fly, as and when you need it. Nor can you add more processor cores on the fly. With horizontal scale, it's a no brainer but with vertical scale you cannot scale-up temporarily and then scale back down either. This is quite common and desirable for web applications.
I was on BizSpark and I was really grateful for the free trial. However, ultimately I shifted away and I am glad I did, even though C# is awesome and I absolutely love Visual Studio, I moved away for a number of compelling reasons.
You can buy msdn packages for unlimited development use of any OS across VMs. Around £600 / year I believe. More expensive MSDN subscriptions include development OSes, Visual Studio & Sql Server at around £1000 / year.
1. You don't have to buy every version. We just got rid of Windows Server 2008 and moved to 2012 R2. We have 5 Windows 2003 Servers that are still supported hanging on until next year.
2. You can hire Visual Studio per month so scaling the team up and down is cheap and easy. Also 100% honestly, VS is pretty cheap compared to the cost of staff and kit. In fact we use VL (renting software) so we get any version we want of any dev tool, server platform etc.
3. It's fairly easy managing 100+ (well 154 but close enough) Windows machines. We do it fine with two guys and a network monkey. Powershell DSC and they are quite scriptable and easy to deploy these days. We build WIM files.
4. You shouldn't plan to add RAM or cores on the fly. Stick a duff RAM stick in and what happens? Scale up/down in/out: depends who you host with.
Really curious: what did you move to (both language and IDE-wise)?
I doubt that this type of instant scaling is useful for the typical .NET shops that serve small and medium size businesses and still sell a lot of on-premise solutions and niche-market web apps.
> That's one OS license, one SQL Server license and one Visual Studio license per developer.
Not necessarily. I just worked with a client who had every one of their 10 developers on SQL Server Express (free, with 10GB db limit) and Visual Studio 2013 Express (also free) which is good enough for doing basically anything in the MS universe. The computers they bought came with Windows.
My current client has me working at home where I use my own copy of Windows 2012 R2 + Visual Studio 2013 Pro and SQL Server Web edition which I got all for free via a free WebSpark membership that I signed up for 2 years ago.
MS seems to be getting a lot better about this stuff (dev licenses, free short-term VM images for testing purposes, I don't know what else) but making things cheaper while not relaxing compliance requirements doesn't make it all that much easier.
Jeff Atwood made a GREAT comparison on up-vs-out-scaling some years ago:
Yeah, servers cost money to provision. But then servers always cost money to provision regardless of your stack. Even with license fees, the MS stack wins for me because my entire server farm fits on three boxes instead of the twelve I'd need to get the same firepower on a typical rails stack. The extra few thousand in full price server licenses is just noise because really, we're talking about a business here, not a hobby.
The bottom line is that if your business is going to fail because of a single $5,000 invoice, it's debatable whether it was ever a business to start with. The whole reason we all build SaaS businesses in the first place is the 95%+ profit margins. Fretting about what happens within that 5% that makes up the expense column seems a bit silly.
Even without BizSpark, I bet I'll build the next product on the same stack. The math just plain works out.
We host it all on Mono and Docker on Linux VPSs. We wanted Docker because we wanted PaaS-like deployment but not vendor lock in.
Our team uses Linux, Windows and OSX to work on it. MonoDevelop is a pretty decent tool, on some areas actually better than VS.
We're really happy about the choice. C# is a great language, NuGet beats pip and gem by miles (none of that virtualenv bull). If Mono support is a requirement from the set-out, you don't run into many compat problems.
Perhaps a post on the company blog. It would make great reading here on HN. The community gets to hear about a C#+Postgres+Mono+Docker stack and maybe you'll get some new users for your app or bright people making job inquiries. At the very least, increase brand awareness.
Just don't forget; the company logo on the blog page should link to the app, not the blog.
Server setup: We have 6 docker containers: Postgres, backend, nginx, assets, data and ssh. Backend is a ubuntu+mono+backend_code. Data is the docker data container pattern. Assets is the frontend code (we're an SLA). Ssh is an sshd for looking at the data and logs in the data container.
Nginx and backend talk via fastcgi, one of ServiceStack's apphosts.
Currently it's all on one server, but the code is prepared for scaling and load balancing.
Also using SQL Azure and Redis for data storage/caching - it's been mostly positive so far. The downside with Azure has been the portal/management experience. The current UI is somewhat sluggish and disorganized, and the new portal tile UI is extremely confusing to navigate.
Overall it's been a pretty pleasant experience (maybe minus dealing with web.config files).
Seriously. I don't know what they were thinking. I don't mind the current portal. It does take seem sluggish with all the ajax/loading but at least I know where everything is. The new portal needs to be completely redone.
It's hard to say no when they give you so much free stuff for a startup-style business.
For the most part, the stuff that runs on Azure doesn't have a discrete licensing cost. You pay the hourly rate for the resource.
free servers are free servers.
From developer's perspective it's two modern mainstream languages - F#, C# - that can be used on Windows, Azure, Mac, iOS, Android, WP as a well-defined workflow.
The best IDE - Visual Studio + ReSharper + soon Roslyn. Second best ecosystem of libraries, that is picking up.
There are some missing bits that are coming: docker and containers, dynamic compilation (of static typed code) when building web sites.
It might not be the best choice in individual "disciplines", but it's the best overall.
I've found the best value/growth combination to be Linux, Mono, and F#. If I need a database I'll use something free or open source.
Right now I think folks are making a mistake if they decide to go "all Microsoft" or "no Microsoft". It's much more nuanced than that.
Bizspark is also great! $150 per month of free credit really helps - especially at the early stages.
* ASP.NET for web with SQL Server + EF for the database and data access
* Xamarin for mobile and OData for data access (back to our ASP.NET sites)
The CLR is great because it's always been fast enough for our applications and every so often you get something for nothing. For example, CLR 4.5 brought better GC and RyuJIT is bringing better code generation. NET Native might give us even better performance.
LINQ is amazing if you know your IQueryables from your IEnumerables and the difference between an Expression<Func<T, bool>> and a Func<T, bool>.
Hosted in a private datacentre, not using Azure but we've got our eye on it (provisioning resources in the Azure Portal seems to get easier and easier than doing it ourselves).
New Relic for monitoring the web sites.
After that I'd say Xamarin is letting the team down. Every second or so release of theirs fixes two bugs but adds another one.
Keeping an eye on Application Insights. It looks like it integrates nicely with Visual Studio Online, and VSO/TFS actually looks quite nice in parts (e.g. Bamboo for CI is largely geared towards Java apps not .NET but TFS' Release Management looks promising). I have my eye on VSO/TFS too but right now it looks like pros do not outweight the cons vs. our Atlassian applications.
If you've got enough processing power and memory and a fast SSD to run Visual Studio + ReSharper it's coding heaven.
The good will become perfect soon enough:
I love AI. I hate that they don't even provide a clue to what the pricing model for this will be though.
Personally, I've always enjoyed C#, so it's been a pretty positive experience on the whole.
There was an announcement of Clojure in Unity3d  a couple weeks ago which is ClojureCLR.
Only downside is that for something like a customer-facing portal, the identity-management is tied to Windows-specific features, for now - so have had to build a custom solution for this that runs on Linux. Waiting for ASP .NET vNext and its Identity solution when it becomes slightly more mature.
EDIT: Also evaluating a move to Azure. Primary driver (for choosing the .net stack) has been prior experience with C#, .NET and Visual Studio
All .net/sql server, jquery on the front end with websockets.
Like others here who get set up on a MS stack, everything integrates very easily. Visual Studio IMO is still one of the best IDEs out there for what it does, and having that supported by a good CI system means very little is left to do except code.
I want to praise Microsoft's BizSpark program. I was accepted a few months ago after I submitted my plan for a SaaS product. I am using Ubuntu servers on Azure, and Azure works well - fewer services than AWS, but a comparable developer's experience to AWS and Google's offerings. I definitely recommend looking into BizSpark if you are interesting in developing an SaaS business. Since I am using Linux, Haskell, and Ruby I am not really locked into using Azure, but if my SaaS product is successful, I will probably keep using Azure in gratitude for $150/month of free Azure use for up to 3 years.
In general my attitude about Microsoft improved a lot earlier this year after trying their Office 360 product that works on all of my devices (MacBook, iPad, and Android phone). I like the direction the new CEO is taking the company.
There are others as well (I just don't recall their names right now) -- take a look on http://fsharp.org for more links.
And https://www.calendartree.com/ (startup by Dilbert's Scott Adams + others) are C# / SQL site, on their own servers.
Reason for using it was tooling, hosting and familiarity. I enjoy being able to being able to fully debug the code and step through things. Makes things enjoyable, to say the least. I love C# and use it for everything I do.
Only concern is the steep SQL pricing for the new tiers and the occasional blips as mentioned by siganakis. However, support is fast and helpful.
Shameless plug: we're hiring. We're distributed/remote between SF, SLC and LA. Ping me for futher details.
Disclaimer: former MS dev.
We started with Bizspark as well, now we graduated and license cost is not much really. We also rely on MS stack for many things other than source code repository / CI etc.
Mostly ASP.NET MVC, but we also use Akka.NET for a lot of our internal services.
For C#, not interested: The syntax borrows from the
deliberately idiosyncratic syntax of C, and
the Visual Basic syntax is more verbose,
easier to learn, teach, and read, and, thus,
less error prone, is as a programming language
in its semantics, features, and functionality
essentially equivalent to C#, that is,
the difference is mostly just syntactic sugar,
and apparently is about as
good as C# or anything else at getting at
the CLR and the .NET Framework.
people back to Backus have been working on
functional programming, and, for
more than just curiosity, and certainly
for the work of our start-up, we're not
impressed or interested, in that or
logic programming, rule-based programming,
or standing on our heads in the corner
programming, etc. -- sorry 'bout that;
we can agree to disagree, and YMMV.
We're quite pleased with Visual Basic
.NET. A hat tip to the Microsoft
Visual Basic group.
Main gripe: For
the documentation, actually for all the
documentation on MSDN, especially for
SQL Server, please, please,
please, get a least a C in
technical writing 101. Our biggest problem
in our start-up, far and away, much, much,
much more difficult than everything else
combined, literally, is the low quality of
the technical writing at MSDN.
some for us, but so far we have yet to write
a single line of it and will delay doing
so as long as we can, hopefully forever.
For ASP.NET, we are using just the basics,
that is, with
We type into our favorite text editor and
make no use of an IDE nor do we want to.
For developing code for Web pages,
ASP.NET and Visual Basic work nicely together
-- it works fine. Actually, it's nice.
For a session state store, we wrote our
own using just instances of one class,
de/serialization, TCP/IP raw sockets,
and two collection classes. On an 8
core processor at 4.0 GHz, our session
state store should be good for 11,000+
transactions a second; transactions
is appropriate because the code is
single threaded. Yes, on an 8 core
processor we would run several
instances with sharding.
As we go live, our server will be just
a single full tower case with an 8
core processor, 32 GB of main memory,
some SSDs for a special purpose
particular to our start-up,
and some disks, with Windows
Server, SQL Server, IIS, ASP.NET, our
Web pages, our session state store,
and some specialized back-end servers.
The software and architecture are
scalable via just simple sharding.
We have little or no intention of using
the cloud soon or ever.
As we go live and grow,
after the issue of the MSDN documentation,
no doubt, and closely related to the
documentation issue, our biggest concern
will be Windows Server, SQL Server, and IIS
installation, configuration, updates,
security, monitoring, management,
performance, administration, reliability,
backup, and recovery -- little things like
those. But from some of the just astounding
successes Microsoft has running server
farms with their software, clearly
terrific results are possible; maybe we
will be calling for some high end technical
That's what we are doing!
What it does: http://bit.ly/1rqJ0NG
Security aspects: http://bit.ly/1tgHUcD
2FA/OTP ideas: http://bit.ly/1sQyQtq
For WCF, LINQ, and WPF, we looked at the
documentation, and it was clear that just
working through the tough parts of the documentation
alone would be much more work than just writing
our code directly. Moreover, any bugs,
security problems, need for updates, etc. of
WCF, LINQ, and WPF we just neatly avoided.
And we saved maybe a lot of address space.
That is, it looked to us like, for us, YMMV,
WCF, LINQ, and WPF are a candle not worth
the match. I can believe that in some
situations those three could have some value;
for us, I don't see the value.
I mean, WCF boils down to TCP/IP raw sockets,
right? Well, the API to raw sockets is
about the easiest I/O one could hope for,
nearly as easy as the ancient Basic statement
PRINT or the same in C, etc. I tried for
several days to make sense out of WCF,
came up with zip, zilch, zero, nichts,
nil, nada, and gave up,
but I learned raw sockets in about three
hours one afternoon decades ago. To heck
For LINQ, that boils down to just SQL, right?
Well, SQL was supposed to be really easy to
use, is profusely documented, and has been,
for decades, has long been an ANSI
standard, has been tested in various ways,
and with astounding severity, around the world,
for decades, right? So, writing some SQL
is not so tough. Getting a connection
string to SQL Server? Yup, that took two
weeks of mud wrestling, but the SQL statements
were easy enough to write. SQL's fine.
Very glad to have it. SQL Server connection
string mud wrestling? Did I mention some
bad documentation? The full power of the
Microsoft Corporation still is unable
to document how to write SQL Server
connection strings. The only way I
learned was via Google searches wherever,
and then I did some organized trials
of essentially all possible combinations
of everything involved until it appeared
to work. I still don't know just why
what worked did and what didn't work
didn't. Cave men with burnt sticks
wrote better documentation on cave
For our SQL Server database schema,
that is, the specifications of
the tables, the columns, etc.,
we have a little file, just awash in
comments, nicely clear and easy to read,
that we run through
the wonderful SQLCMD.EXE -- "Look,
Ma, no GUIs!". And no clicking!
Instead, just type the name of a little
script, and presto, boom, have
a new database created just as
want, nearly no chance of anything
going wrong. It's better than most sex!
For the SQL Server Management Studio, can use
it read only a little and otherwise just
can't make sense out of most of the screens,
windows, sub-windows, check boxes, terminology,
So, for SQL Server administration,
just back to the simplest
little SQL statements, mostly industry standard.
The SQL Server documentation of users, logins,
etc. wins the prize for the worst technical
writing I ever saw. But, do some Google
searches, go to some Web sites, and see
that just some fairly simple SQL statements
are enough for such administration work.
So, get a collection of little SQL files,
run them through the wonderful, fantastic,
we're lovin' it, SQLCMD.EXE, document
the little files and their usage, and, fine --
f'get about Management Studio.
Basically, right, really prefer
command lines to GUIs! GUIs? Far
too difficult to understand,
get to work, script,
document, etc. We'll let the
cognitive psychologists at Xerox
PARC, Steve Jobs, etc. play with
direct manipulation GUIs, and we'll
stay with command lines, scripts
driving command lines, and a text
editor writing the scripts. Sorry
'bout that. To each their own.
If you like GUIs, then go for it.
WPF? Maybe that boils down to Windows
Forms for a native Windows application,
but our only GUI is HTML and CSS
for us. If WPF boils down to HTML elements,
then, sure, we'll just use, directly, HTML elements
and/or the ASP.NET class wrappers for
those elements to make the HTML
programmable, that is, be able to
generate HTML elements via software
that executes for each user
uniquely after they connect.
So, we just saved having to read some
hundreds of Web pages at MSDN, buy
a cubic foot or so of books, do 500
Google searches about details, submit
100 questions to Stack Overflow and
the MSFT fora, etc.
Our code so far is about 30 KLOC
(that is, actual VB statements)
with maybe 60 KLOC of blank lines
and 100 KLOC of comments.
I think I need to look into TCP/IP sockets though. I have a very simple WCF service that accepts a byte array, and returns a byte array. No more, no less. The back-end deserializes the bytes and does what it needs to. WCF configuration is a Night! Mare! I think someone got completely spannered on the dependency injection koolaid. My service is a single method, and I spend more time babysitting XML in Web.Config than I do writing the service...
And yeah, with white space and comments my code base is around 120 LKOC. Just under 140 if I include the Windows Phone client.
So, in a kitchen, I don't want a hot dog
cooker and, instead, can, depending on
what I want, just use a pot and with
water and boil the hotdogs, use a skillet
with some oil and fry the hot dogs, or
go on the back porch, build a charcoal
fire, and broil the hotdogs. Besides, for
a hotdog cooker, how to clean the darned
thing! And have to get it out of storage,
set it up, etc. No thanks.
For cutting onions, use my best three tools
in a kitchen, a good French chef's knife,
a cutting board, and my hands. Works great
also for potatoes, carrots, cabbage, garlic,
and much more.
In a shop, have saws, hammers, files, wrenches,
screwdrivers, a drill plate, drill bits,
a nice, variables speed, reversible 3/8"
electric (not battery powered -- don't want
to mud wrestle with batteries that run down,
need charging, go dead, wear out, get weak, need to
be replaced, etc.) drill. Also
socket wrenches, etc. But lots of automatic
tools, e.g., a screwdriver that has a spiral
shaft that push on to turn a screw, don't want.
To sharpen a kitchen knife? Sure, just use
whatever grit size sand paper, usually
silicon carbide, want, put the paper flat
on the cutting board, and sharpen away.
Simple tools, that do simple things, and are
reliable and easy to understand. Then, build
more on those.
So, just use TCP/IP raw sockets and build more
For a session state server, use TCP/IP raw sockets,
de/serialization, and two collection classes.
Piece of cake, no use of CONFIG whatever, no
use of XML, only a few pages of documentation,
lots of good, old code samples, rock solidly
reliable ("TCP/IP sockets are the workhorse of
all of the Web and Internet")
etc. By far the
...if not, would you mind letting us email your developers to try and save them from career suicide?
I'll let them all know that they have a
choice: (1) Make money doing the work to
make our software and Web site good,
our users happy, and our start-up successful
or (2) grow their resumes for jobs
With (1), they get to help build something
important. With (2) they get to add buzz
words of questionable utility for unknown
projects of questionable value. YMMV.
I'd say obvious ... is obvious
Just in case you guys aren't clear on the intent of the post, another submission by this user talking about cats breaking Firefox:
Ah, good to get some laughs.
But if your kitty cat, puppy dog, or baby
walks across your keyboard, or if something
falls on your keyboard, etc., and suddently
Firefox closes after 10 seconds, and keeps
as you keep restarting Firefox,
no matter what,
very well like my post and its solution.
Or, you can do a Google search and find that
the symptoms are common, and find that the
recommended solutions range from reinstall
your anti-virus software, install a new
version of Firefox, reinstall device drivers,
..., buy a new keyboard, give up your kids,
divorce your spouse, buy a new computer,
submit to full castration,
move to the woods in northern Canada, or worse.
Instead, just use the solution I found.
Piece of cake.
It's easy to suspect that my simple discovery
will solve a major fraction of all the
instances of that nasty, fairly common
problem, hilarious or not.
Exercise: Before reading my solution, guess
what it is! Record your guesses here before
reading my solution.
Hilarious? Maybe. Obvious? No. Useful?
Yes. Difficult? No. Fast? Yes. Easy?
Definitely. Better than the suggested
alternatives? By some light years.
Either way: LOL
No, what I wrote is what we're doing. Maybe
there's some humor in there somewhere, but
it's all simple and obvious and makes good
sense to us.