Scott Hanselman

This last week Joseph Phillips tweeted that he was going to his first big tech conference and wanted some tips and suggestions. I have a TON of tips, but I know YOU have more, so I retweeted his request and prompted folks to reply. This was well timed as I had just gotten back from OSCON and BUILD, two great conferences.

The resulting thread was fantastic, so I've pulled some of the best recommendations out. As per usual, the Community has some great ideas and you should check them out!

• @saraford - Whenever you get a biz card write down why you met them or what convo was about. It might seem obvious at time but you wont remember at home
• @arcdigg - Meet people and speakers. Tech is part of your success, but growing your network matters too. Conf can give you both or not. Up to you!
• @marypcbuk - if approaching people is hard for you, just ask 'what do you work on?'
• @ohhoe - don't be afraid to introduce yrself to people! let them know its yr first conference, often people will introduce you to other people too :)
• @IrishSQL - connect with a few attendees/speakers online prior to event, and bring plenty of business cards. When u get one, write details on back
• @arcdigg - Backpack and sneakers beat cute laptop bag and heels (ed: dress comfortably)
• @scribblingon - You might feel left out & think everyone knows everyone else. Don't be afraid to approach people & talk even if seems random sometimes :) If you liked someone's talk, strike a convo & tell them that!!
• @arcdigg - Plan session attendance in advance, have a backup in case the session is full.
• @jesslynnrose - Reach out to some other folks who are using the hashtag before you get there, events can be cliquey, say hi and make friends before you go!
• @thelarkinn - Never feel afraid to say hi to maintainers, and speakers!!!! Especially if you want to help!
• @everettharper - Pick 3 ppl you want to meet. Prep 1 Q for each. Go early, find person #1 in the 1st hr before crowds. 1/3 done = momentum for rest of day!
• @jorriss - Meet people. Skip sessions. You'll get more from meeting and talking with people then sitting in the sessions. #hallwaytrack
• @stabbycutyou - Leave room in your schedule, Meet people, Eavesdrop on hallway convos, Take notes, Present on them at your job
• @patrickfoley - Don't forget to sleep. Evidence that long-term memories get "written" then
• @david_t_macknet - Drinking will not help you remember it better or have a better time mingling. Most of us are just as introverted & the awkwardness fades.
• @carlowahlstedt - Don't feel like you have to go to EVERY session.
• @davidpine7 - Try your best to NOT be an introvert -- in our industry that can be challenging, but if you put yourself out there...you will not regret it!
• @frontvu - Don't rely on the conference wifi
• @shepherddad - Put snacks in your bag or pocket.
• @sod1102 - Find out if there will be slides (and even better!) video available post conference, then don't worry about missing stuff and relax & enjoy
• @rnelson0 - Take notes. Live tweet, carry a notebook, jot it all down at 1am before sleeping, whatever method helps you remember what you did.
• @hoyto - Sit [at] meal tables with random people and introduce yourself.
• @_s_hari - Ask speaker when *not* to use product/methodology that they're speaking on. If they cannot explain that, then it's just a marketing session
• @EricFishor - Don't be afraid to discreetly leave or enter an on going session. It's up to you to seek out sessions that interest you.
• @texmandie - If you get to meet and talk to your heroes, don't freak out - they're normal people who happen to do cool stuff
• @wilbers_ke - Greatest connections happen in the hallways, coffee queue and places with animated humans. Minimize seated conference halls
• @CJohnsonO365 - CLEAR YOUR SCHEDULE. Don’t try to get “regular” work done during the conference— you’ll end up missing something important!
• @g33konaut - Tweet with the conf hashtag to ask if people wanna meet and talk or hangout after the conference, also follow the hashtag tweets to find ppl. Don't sweat missing a talk, meeting people and talking to them is always better than than seeing a talk. Also the talks are often recorded
• @foxdeploy - Who cares about swag, it's all about connections. Meet the people who've helped you over the years and say thanks.
• @jfletch - Ask people which after parties they are attending. Great way to find out about smaller/more interesting events and get yourself invited!
• @marxculture - The Law of Two Feet - if you aren't enjoying a session then leave. Go to at least one thing outside your normal sphere.
• @joshkodroff - Bring work business cards if you're not looking for a job, personal business cards if you are.
• @benjimawoo - Go to sessions that cover tehnologies you wouldn't otherwise encounter day to day. Techs you don't use in your day job.

Fantastic stuff. You'll get more out of a conference if you say hello, include the "hallway track" in your planning, stay off your phone and laptop, and check out sessions and tech you don't usually work on.

What are YOUR suggestions? Sound off in the comments.

Sponsor: Did you know VSTS can integrate closely with Octopus Deploy? Watch Damian Brady and Brian A. Randell as they show you how to automate deployments from VSTS to Octopus Deploy, and demo the new VSTS Octopus Deploy dashboard widget. Watch now!

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

About   Newsletter

Hosting By

The BUILD Conference was lovely this last week, as was OSCON. I was fortunate to be at both. You can watch all the interviews and training sessions from BUILD 2017 on Channel 9.

Here's a few sessions that you might be interested in.

Scott Hunter, Kasey Uhlenhuth, and I had a session on .NET Standard 2.0 and how it fit into a world of .NET Core, .NET (Full) Framework, and Mono/Xamarin.

One of the best demos, IMHO, in this talk, was taking an older .NET 4.x WinForms app, updating it to .NET 4.7 and automatically getting HiDPI support. Then we moved it's DataSet-driven XML Database layer into a shared class library that targeted .NET Standard. Then we made a new ASP.NET Core 2.0 application that shared that new .NET Standard 2.0 library with the existing WinForms app. It's a very clear example of the goal of .NET Standard.

Then, Daniel Roth and I talked about ASP.NET Core 2.0

Maria Naggaga talked about Support for ASP.NET Core. What's "LTS?" How do you balance purchased software that's supported and open source software that's supported?

Mads Torgersen and Dustin Campbell teamed up to talk about the Future of C#!

David Fowler and Damian Edwards introduced ASP.NET Core SignalR!

There's also a TON of great 10-15 min short BUILD videos like:

• Get started with Unity and Visual Studio for Mac
• .NET Core and Visual Studio for Mac
• Windows High DPI Improvements for Desktop
• Miguel de Icaza and Scott Hunter on .NET Standard 2.0, UWP Support, and UI Futures on CH9

As for announcements, check these out:

• Announcing EF Core 2.0 Preview 1
• Announcing .NET Core 2.0 Preview 1
• Announcing ASP.NET Core 2.0.0-Preview1 and Updates for .NET Web Developers
• Visual Studio 2017 Tools for Azure Functions
• A Lap Around Python in Visual Studio 2017
• Unity game development with Visual Studio for Mac
• Why you should use F#
• Announcing F# 4.1 and the Visual F# Tools for Visual Studio 2017
• All Things Mobile at Microsoft Build

And best of all...All .NET Core 2.0 and .NET Standard 2.0 APIs are now on http://docs.microsoft.com at https://docs.microsoft.com/en-us/dotnet

Enjoy!

Sponsor: Test your application against full-sized database copies. SQL Clone allows you to create database copies in seconds using MB of storage. Create clones instantly and test your application as you develop.

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

About   Newsletter

Hosting By

Tons of great announcements this week at the BUILD conference. I'll slowly blog my take on some of the cooler features, but for now here's a rollup of the major blog posts for developers:

• Announcing .NET Core 2.0 Preview 1
• Announcing ASP.NET Core 2.0 Preview 1 for .NET Web Developers

You can download and get started with .NET Core 2.0 Preview 1 right now, on Windows, Linux and macOS:

• .NET Core 2.0 Preview 1
• Visual Studio 2017 Preview 15.3 for Windows - installs side by side with your existing released version
• Visual Studio Code

If you already have .NET Core on your machine, you'll already be able to type "dotnet --version" at the terminal or command line. Go ahead and try it now. Mine says:

C:\Users\scott> dotnet --version
2.0.0-preview1-005977

Remember on Windows you can check out c:\program files\dotnet\sdk and see all the SDK versions you have installed:

Typing dotnet will pick the most recent one...but it's smarter than that. Remember that you can set the current SDK version with a global.json file. Global.json's presence will override from the folder its in, all the way down.

If I make a folder on my desktop and put this global.json in it:

{
  "projects": [ "src", "test" ],
  "sdk": {
    "version": "1.0.3"
  }
}

It will force my dotnet runner to use the .NET Core SDK version I asked for. That "projects" line isn't needed for the versioning, but it's nice to be able to select what folders have projects inside.

C:\Users\scott\Desktop\test> dir
 Directory of C:\Users\scott\Desktop\test
05/11/2017  09:22 PM    <DIR>          .
05/11/2017  09:22 PM    <DIR>          ..
05/11/2017  09:23 PM                45 global.json
               1 File(s)             45 bytes
               2 Dir(s)  85,222,268,928 bytes free

C:\Users\scott\Desktop\test> dotnet --version
1.0.3

At this point - with a valid global.json - making a new project with dotnet new will make an app with a netcoreapp1.x version. If I move elsewhere and dotnet new I'll get a netcoreapp2.0. In this example, it's the pretense of that global.json that "pins" my SDK version.

Alternatively, I could keep the dotnet.exe 2.0 SDK and install 1.x templates. This would mean I could create whatever I want AND pass in the version.

First I'll add the 1.x templates into my 2.0 SDK. This just needs to happen once.

dotnet new -i Microsoft.DotNet.Common.ProjectTemplates.1.x::1.0.0-*

Now, even though I'm "driving" things with a .NET Core 2.0 SDK, I can pass in --framework to control the project that gets created!

C:\Users\scott\Desktop\test> dotnet new console -o oneone --framework netcoreapp1.1
The template "Console Application" was created successfully.

C:\Users\scott\Desktop\test> dotnet new console -o twooh --framework netcoreapp2.0
The template "Console Application" was created successfully.

I can make libraries that target .NET Standard like this, passing in 2.0 or 1.6, or whatever netstandard I need.

C:\Users\scott\Desktop\lib> dotnet new classlib --framework netstandard2.0
The template "Class library" was created successfully.

There's two options that are not exactly opposites, but they'll give you different levels of control, depending on your needs.

• You can control your SDK versioning folder by folder with global.json. That means your project's directories are "pinned" and know what SDK they want.
• When you type dotnet new using a pinned SDK, you'll get the new project results for that pinned SDK. Typing dotnet run will do the right thing.
• You can pass in --framework for templates that support it and dotnet new will create a template with the right runtime version. Typing dotnet run will do the right thing.

This is .NET Core 2.0 Preview 1, but you should be able to install it side by side with your existing apps and have no issues. If you know these few internal details, you should be able to manage multiple apps with multiple versions without much trouble.

Sponsor: Test your application against full-sized database copies. SQL Clone allows you to create database copies in seconds using MB of storage. Create clones instantly and test your application as you develop.

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

About   Newsletter

Hosting By

I was getting frustrated with the speed (or lack of) of Windows Search within Windows 10 lately. I went over to my Indexing Options - just hit Start and type Indexing Options - and was surprised to see that there was over 1.5 MILLLION items indexed! That seems like a big number. Why so large?

I checked my "index these locations" list and didn't see anything weird, but I did note that Indexing does include c:\users\YOURNAME by default. That seems reasonable, because it is reasonable.

However, I also noted that I had a LOT of ".folders" (dot folders) under my C:\users\YOURNAME folder adding up to a few gigs of config text files, caches and general crap.

I was able to significantly lower the number of items indexed from over a million to a reasonable 215k items just by excluding (un-checking) folders that I knew didn't matter to me as much.

Go to Indexing Options and click Modify:

Go to your C drive (or wherever ~\YOURNAME is) and go to your top level User folder. I unchecked a bunch of the stuff that didn't matter to me.

For average users this won't matter, but for developers who install a bunch of utilities, have their Dropbox or OneDrive in the c:\users folder, a 5 min audit of your indexed files can give your Indexed Files a nice refresh.

Sponsor: Check out JetBrains Rider: a new cross-platform .NET IDE. Edit, refactor, test, build and debug ASP.NET, .NET Framework, .NET Core, or Unity applications. Learn more and get access to early builds!

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

About   Newsletter

Hosting By

I got an interesting question today. This is actually an extremely common one so I thought I'd take a bit to explore it. It's worth noting that I don't know the result of this blog post. That is, I don't know if I'll be right or not, and I'm not going to edit it. Let's see how this goes!

The individual emailed and said they were new to Azure and said:

Question for you.  (and we may have made a mistake – some opinions and help needed)
A month or so ago, we setup a full up Win2016 server on Azure, with the idea that it would host a SQL server as well two IIS web sites

Long story short, they were mired in the setup of IIS on Win2k6, messing with ports, yada yada yada. '

All they wanted was:

• The ability to right-click publish from Visual Studio for two sites.
• Management of a SQL Database from SQL Management Studio.

This is a classic "lift and shift" story. Someone has a VM locally or under their desk or in hosting, so they figure they'll move it to the cloud. They LIFT the site as a Virtual Machine and SHIFT it to the cloud.

For many, this is a totally reasonable and logical thing to do. If you did this and things work for you, fab, and congrats. However, if, at this point, you're finding the whole "Cloud" thing to be underwhelming, it's likely because you're not really using the cloud, you've just moved a VM into a giant host. You still have to feed and water the VM and deal with its incessant needs. This is likely NOT what you wanted to do. You just want your app running.

Making a VM to do Everything

If I go into Azure and make a new Virtual Machine (Linux or Windows) it's important to remember that I'm now responsible for giving that VM a loving home and a place to poop. Just making sure you're still reading.

NOTE: If you're making a Windows VM and you already have a Windows license you can save like 40%, so be aware of that, but I'll assume they didn't have a license.

You can check out the Pricing Calculator if you like, but I'll just go and actually setup the VM and see what the Azure Portal says. Note that it's going to need to be beefy enough for two websites AND a SQL Server, per the requirements from before.

For a SQL Server and two sites I might want the second or third choice here, which isn't too bad given they have SSDs and lots of RAM. But again, you're responsible for them. Not to mention you have ONE VM so your web server and SQL Server Database are living on that one machine. Anything fails and it's over. You're also possibly giving up perf as you're sharing resources.

App Service Plans with Web Sites/Apps and SQL Azure Server

An "App Service Plan" on Azure is a fancy word for "A VM you don't need to worry about." You can host as many Web Apps, Mobile Apps/Backends, Logic Apps and stuff in one as you like, barring perf or memory issues. I have between 19 and 20 small websites in one Small App Service Plan. So, to be clear, you put n number of App Services as you'd like into one App Service Plan.

When you check out the pricing tier for an App Service Plan, be sure to View All and really explore and think about your options. Some includes support for custom domains and SSL, others have 50 backups a day, or support BizTalk Services, etc. They start at Free, go to Shared, and then Basic, Standard, etc. Best part is that you can scale these up and down. If I go from a Small to a Medium App Service Plan, every App on the Plan gets better.

However, we don't need a SQL Server, remember? This is going to be a plan that we'll use to host those two websites. AND we can use the the same App Service Plan for staging slots (dev/test/staging/production) if we like. So just get the plan that works for your sites, today. Unlike a VM, you can change it whenever.

SQL Server on Azure is similar. You make a SQL Server Database that is hosted on a SQL Server that supports the number of Database Throughput Units that I need. Again, because it's the capital-C Cloud, I can change the size anytime. I can even script it and turn it up and down on the weekends. Whatever saves me money!

I can scale the SQL Server from $5 to a month to bajillions and everything in between.

What the difference here?

First, we started here

• VM in the Cloud: At the start we had "A VM in the Cloud." I have total control over the Virtual Machine, which is good, but I have total control over the Virtual Machine, which is bad. I can scale up or out, but just as one Unit, unless I split things up into three VMs.

Now we've got.

• IIS/Web Server in the Cloud: I don't have to think about the underlying OS or keeping it patched. I can use Linux or Windows if I like, and I can run PHP, Ruby, Java, .NET, and on and on in an Azure App Service. I can put lots of sites in one Plan but the IIS publishing endpoint for Visual Studio is automatically configured. I can also use Git for deployment as well
• SQL Server in the Cloud: The SQL Server is managed, backed up, and independently scalable.

This is a slightly more "cloudy" of doing things. It's not microservices and independently scalable containers, but it does give you:

• Independently scalable tiers (pricing and CPU and Memory and disk)
• Lots of automatic benefits - backups, custom domains, ssl certs, git deploy, App Service Extensions, and on and on. Dozens of features.
• Control over pricing that is scriptable. You could write scripts to really pinch pennies by scaling your units up and down based on time of day or month.

What are your thoughts on Lift and Shift to IaaS (Infrastructure as a Service) vs using PaaS (Platform as a Service)? What did I forget? (I'm sure lots!)

Sponsor: Check out JetBrains Rider: a new cross-platform .NET IDE. Edit, refactor, test, build and debug ASP.NET, .NET Framework, .NET Core, or Unity applications. Learn more and get access to early builds!

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

About   Newsletter

Hosting By