Welcome to the Reset

TL;DR Edition: Create a new site in the new azure portal, Get IIS error. Create new site in classic portal, works fine.

Full Post: When I first started using azure - I loved it and I mean LOVED IT! As a web developer, I could spin up a site, deploy my code to it, it just worked and didn’t have to configure the server at all. At my previous employment, I pushed hard to try and get them to use azure, within my first couple of months I used my personal account for demos since the staging environment was just configured in an annoying way. A year later once I finally was able to get a client to use azure, I felt it was a start of something big - First it was much cheaper compared to other hosts we were using, deployment slots allowed an easy roll back in the event someone pushed before they got the green light to do so (I made sure that when I wrote the instructions on deploying to azure that the only way to do so was to create a deployment slot - so anyone who was new that never used azure would think that was the only way and if someone knew how to use azure that got hired after me, well I think they would agree to my process), and a big plus was I didn’t have to worry about half of the security nightmare that 2014 was or having to hope a server completes a reboot at 4am.

During that time span, I got close to the new portal in azure, the classic portal was slow and wasn’t as easy as the new one. I kept a close eye on the Azure Friday videos and pretty much had nothing negative to say. Fast Forward to the current day - left my previous employer and now at my current employer we pretty much are all azure, the new portal changed a bit (not much of look and feel, but more of wording - I’m sure that obviously it had major changes not only in the back end but also the front end), suddenly the classic portal must of had some improvements because it was quicker and I started notice things which was a bit odd.

At some point, we received word from our remote team that they had trouble accessing the site, which at first we assumed it was internet issues because they were having internet issues a few days previous and azure wasn’t reporting any issues and then the next thing we knew, we had issues. We would type the domain, hit enter and nothing. Visit another site, instant load. Tried stopping the site, starting it back up. No luck. Checked error logs, nothing stood out. Tried a different client on azure, loads right up. Tried a different deployment slot thinking maybe someone pushed a bad config or code. No luck. How is it that a site that is still under development, that is really only being hit by a small team not respond but HaveIBeenPwned handles an average of 8.5k simultaneous sessions over a period of an hour without any hiccups?

So after a few hours of debugging, making changes to the site we are having issues with and deploying it, one of the developer’s on the team get’s a genius idea - he spins up a brand new website, creates a new default asp.net mvc project, makes no changes to the code, no database connection used, just pretty much the boilerplate mvc site visual studio creates, publishes to azure and hang.... The brand new site didn’t have any configuration changes, just whatever was enabled by default when creating a new azure site, default boilerplate asp.net mvc project and it just doesn’t work. Check azure status - no issues, but clearly it is an azure issue. So now we start trying to figure out what could be the problem - while also trying to explain to those who aren’t developers that the problem isn’t us but azure... which you can see their confused face ask they go “but isn’t azure the cloud? Isn’t it suppose to scale if there is a heavy load, etc?” It’s difficult to try and explain to them that it clearly isn’t the project’s code but something wrong with something with azure without making them start doubting azure - because the last thing you want is someone in a management position that isn’t technical and suddenly is pushing to avoid using azure after X amount of time of convincing them.

Somehow we throw out the idea of maybe it’s because of the new portal, maybe we should try creating the new site with the classic portal. At this point we are clearly just throwing ideas out into the open because the remote team lost a day of work and we had lost half a day trying to debug this issue. So we create a new site in the classic portal, push the default asp.net mvc project to it and loads super quick, we then delete the default asp.net mvc project, deploy the current project we were working on to it and loads super quick, no problems.

End of story right? Not so much, a month or two later we attempt to deploy another site which a different developer worked on who wasn't present at the time. Builds successfully, works locally, publishes with no issues, attempt to view it on azure - no such luck. We keep getting an IIS issue, which makes you wonder - why the fuck am I getting an IIS error on what should be a "simple" web site? Is something wrong with a configuration somewhere? Did we miss something in the web.config transform? One hour passes... then another... it's past 6pm now and I ask "How was this site created?" which I get a response of "Logged in, Create a new site" from coworker.

Me: Was it via the new portal? Coworker: Yeah. Me: I bet you that's why you are getting that IIS error. If you create it in the old portal I bet you it will work.

He deletes the site, creates it in the classic portal and deploys - during this time span I tell him if it works I'm going to have to blog about this if it 'fixes' the issue. I could of been wrong and honestly I wish I was. - Deployment finishes, no iis error, works just fine. We both laugh and then go "We are so screwed"

I know Microsoft has some talented developers and I mean fucking talented! I can't deny that at all - not only that but they have to have it rough depending on what team you are on. I've been a fan of what they've done with azure. The edge team has done great work so far when it comes to my default browser choice at home. Visual Studio tends to have it's quarks after a brand new release which I can understand and it generally gets better overtime. Skype for business.... well I look forward to the day it remembers my password when I click that damn remember my password checkbox but that's a story for another day. ;)

Backstory

So I ran into an interesting issue in which I was using a third party plugin that generated markup around an input field which it hid and also handled the validation of the input. A custom drop down would be a good example of this third party plugin.

Example of the markup (artist rendition): <div class=“wrapper”><span class=“fake-input”></span><input style=“display:none;” ...etc etc /></div>

Now normally you wouldn’t think this would be an issue, since it was handling the validation as well so it wasn’t like I was having to mix different plugins and they weren’t playing nice, how wrong I was in assuming this.

The validation class of if it was successful or unsuccessful wasn’t being added to the wrapper or even the fake-input but to the input... which was hidden. Had the hidden input been placed before the fake-input which was visible it wouldn’t have been such a big deal because I could of used the adjacent selector and style the fake-input however in this case I wasn’t able to do so.

I stumbled upon a blog post by Shan Inman on CSS Qualified Selectors which could solve my issue since then all I would need to do would be something like this: .wrapper < .invalid{border:1px solid red;} assuming that the .invalid class was what was being added to the hidden input. Now that blog post was written in 2008 and after reading some of the comments I noticed Eric Meyer states that it's “Also known as "parent Selectors". ” which seemed to have been brought up multiple times in the past and shot down. Five years later a blog post appears stating how Parent Selectors are appeared in the W3C draft of selectors level 4 for css. 

A New Hope

Now that it’s 2015, I thought to myself - “It’s been two years - I wonder if any browser vendors started supporting it or if there are any polyfills...”. I did discover where someone created a CSS4 Parent Selector polyfill but I was wanting to avoid having to write any JavaScript, on that note - it looked that the polyfill was also deprecated and then I discovered that it was also removed from the specification. :(

I can understand that a parent selector for the most part doesn’t seem like a common use case, and due to the concerns for performance it probably should be used with caution (had it been created) but all I can think is “well damn”.

OneNote for Interaction Designers, Part 2 - Nuts and Bolts: https://medium.com/@christophernoessel/onenote-for-interaction-designers-part-2-874cadc88f83

For the longest time (and even still to this day) I don't consider myself an “expert” - I mean how can someone be an “expert” in the web industry when frameworks are born every 15 minutes.

ex·pert.

NOUN 1.a person who has a comprehensive and authoritative knowledge of or skill in a particular area

So in the web industry there are many who are considered experts, and even within the city are you in there are people who are also considered experts. Now is where it get's scary - you are the expert when it comes to your place of employment.

You could be the intern at a company and be the only one who knows how to use Photoshop (even if it's super basic) then you are the expert at using Photoshop, which might make you the expert for that company’s social media (even if you have no idea what you are doing) - because in someone’s eyes you know how to use Photoshop and can make and update the Facebook banner.

When I was at a Local B2B advertising agency I didn't realize this completely because there were multiple disciplines within the agency, you had SEO, UX, Creative, Development, etc. So when someone might be consulting you from a different discipline it was because you were viewed as the expert in that topic (or because the other person they were looking for wasn't available and your expertise would do) and on the flip side - you would do the same to whoever was asking you for some sort of information, feedback, etc.

It get's a bit scarier when you start getting thrown on calls with clients / facing client teams however I think that's because at some point someone might call bullshit even if they are completely wrong. On the flip side when clients & client facing teams start trusting you and looking for your input, then it's not that bad.

But that was when I was at a large company, so when you get in a much smaller team where you don't know their skill sets and they don't know yours - now is where it get's tricky. In a larger company, I might not have to give much input on a typeface - but on a much smaller company, that input might be critical - such as 'is it websafe?’ and then once you've explained what a websafe font is and why it matters, now you might be asked to make a recommendation or not.

In my case, it was that exactly - Ask the question, explain what it meant and why it mattered, provide information to back up the explanation and could be used as a reference in the future. Once a list of possible choices were presented, noticing a coworker possibly overwhelmed and then casually throwing out the comment of ‘I would pick one of these four or three’. From that point, it's up to them to either say well which one would you pick as the final or decide how they want to handle the final selection.

I can’t remember when I first learned about impostor syndrome, but when I did learn about it I believe my thoughts at the time were something along the lines of “Great, so now I know what to call how I felt a year or so ago and good to know I wasn’t alone”. Imposter Syndrome sucks - for the longest time I put myself down and would say that everything I wrote was a hack. Now to some extent that might be true - I've written my fair share of hacks - I've done things I'm not proud of - but I know everyone has and if you are the one lucky person who never has had to write a nasty hack at 2am to meet a deadline knowing that if you don't that you might fall asleep driving home and that everything you've ever wrote has had enough padding to the timeline to cover all the scope that was introduced before any developer even was involved then good for you.

In 2013 - I met with a recruiter as I was looking for a new job and here I was during my lunch break putting myself down when I should of been doing my best selling myself. Luckily the recruiter was awesome, didn't dismiss me right there and gave me some advice to make a list of accomplishment. I spent the weekend creating that list and it really helped me in pulling out of the slump I was in. I felt that within 3 years I accomplished a lot and was proud of it.

Allow me to paint you the picture of the web as I knew it when I was a child - I would connect to the internet by clicking on a blue triangle labeled America Online that would then tie up a phone line for X amount of hours, where X is the amount of time until someone picked up the phone.

If you wanted a website - it was simple, your friend’s teenage kid could set one up for you. You didn’t even have to buy a domain because you could host your site on Homestead of Geocities. It didn’t matter if you ran out of space, because you could create another account and connect them together as if they were one site with the power of hyperlinks.

Chances are you would forget about your free website after a few years or even after a few weeks until someone reminded you about it, which you would then either update or not.

Fast forward to my high school years - Firefox was the browser to use, google wasn’t evil and you completely trusted and the site to discover cool stuff was called digg.com and new awesome sites were popping up every day. Not everything was connected to the world wide web - so if you wanted to go to a site you bookmarked you would either have to remember what it was or send it to yourself via email... sort of.

Enter web 2.0, everything was connected via APIs - new sites offering free services or you could pay for premium services appeared left and right. So here I was with a small amount of bookmarks, maybe 60 at the time and I wanted to have a way to access these bookmarks from other locations - mostly in highschool and then back home. So I signed up for a free account from a site that wasn’t the Best of or Top Site for XYZ, but it was mentioned in one of those blog posts and had a name I could remember easily so why not put my bookmarks on there. A few months pass and it works great, then the home computer gets struck by lightning twice, having to reinstall Windows XP, etc etc. and I stopped using the browsers bookmarks because this online service works great.

That was most likely the biggest mistake. One day I go to this website and it’s gone. No warning at all, no email saying - we’re shutting down, no blog post saying they has been acquired. Just Gone. At that moment I had to try and recover the sites I now lost and accept the fact that some I would never remember.

I grew up in a time where you could create things quick and while you had a small limit of space and bandwidth you could work with - you could keep adding up accounts as you needed and those services would stick around for years. In fact, by the time I lost my collection of bookmarks within under a year, the sites I created on geocities over five years from that time and hadn’t been updated since a day after they were created still existed.

Sites would start appearing offering a service which was most likely awesome and I wouldn’t sign up for them because I got burned once. Then months later, those sites would be acquired and claim that they would continue to operate as is and then allow you to download your data if you wanted to leave - or worse, just announce that they were shutting down, killing the product and you either had a small time frame to download your data or they wouldn’t allow you to get your data at all.

I signed up for a flickr account but only uploaded a few photos, tried dropbox but I couldn’t trust them to have the only copy of my files. Then in 2009 geocities finally closed their doors. For me this didn’t impact me as much, except for nostalgia factors - however I know it impacted many. How many research papers were lost because the only place they existed and could be cited from were from geocities site created in the 90s and like myself - hadn’t been updated for years but people still used for studies?

I still have trouble trusting web services to some degree - I do use twitter and it started as something that could just be thrown away to well - a way of communicating and getting information that I actually care about however they sort of gained my trust as I can download my twitter archive whenever (granted it’s not perfect and the exception that their support sucks), I do use Github and have no idea how I will handle once it goes away - I mean I wonder how much code will be lost from google killing off Google Code. I sort of use Onedrive, but still don’t trust them with everything and I’m trying to get my OneNote organized slowly but surely. Obviously I’m still using tumblr, however I did run into the panic of how to export my blog posts when Yahoo purchased them and I will mostly at some point mirror my blog post else where one day just in case.

I suppose it’s my own fault for not having a proper back up plan.

A few months back during a gathering over the holidays - I was introduced to a few individuals who were around teenage years I suppose and I remember one teenager who was extremely interested in what I had to say (almost to the point where I started thinking I should hide for a moment of sanity).

The teenager asked what I did for a living which I can't remember how I described it (and I wonder why I didn't just state "I build websites") but the response I got was "Oh like computer engineering? That's cool!" - Not in a sarcastic tone, but the teenager meant it, which I mean the teenager is right. What I do for a living is what I remember being interested in when I was that age - back in the day trying to build websites with geocities.

What I am ashamed of is what I then followed up with. I told the teenager that I wouldn't recommend it at all and in fact I recommend highly against. As I said it - I hated myself.

The truth is - I love what I do, I hate computers but I love being able to solve problems and trying to make things better for people, connecting people, sharing knowledge and ultimately just everything that the internet can empower us to create. The reason I said what I said was when I was younger I didn't know what I know now, in the sense that people suck. Rather - I wasn't aware of how some people live in fear - scared to express their opinions, scared to even share their knowledge because of other people.

Now chances are the teenager like most teenagers thinks XYZ is cool like everything else and chances are this teenager had no interest into entering a career in web development just was interested in talking to an adult who wasn't going to treat them like a little kid, and maybe if they did - maybe they will think I was just in a grumpy mood and ignore what I said - chances are they've forgotten what I said considering it was the first time they met me too.