December 21, 2022 at 3:55 pm
I am working to improve my skills in DevOps and/or cloud and my biggest challenge is this.... My day-to-day org is not very technically mature meaning my direct users mostly cobble data from text files to spreadsheets. Case in point, shifting to PowerBI is a leap for them. For me to take on more cloud-oriented projects will ultimately mean that I would be the one solely supporting them (both the solutions and the required cloud infrastructure). There will be budget considerations if I were to propose adding more cloud services beyond what is currently in scope for this relational data mart. In the long run, I don't know how much practical exposure I can obtain in my current role. Thoughts??
Great question and interesting challenges for you. My main suggestions and thoughts are that a lot my success has been in being effective for my employer/org. That means I not only try to meet their needs, but costs/time/budget/change always come into play.
Not to be too Kung Fu/Karate Kid, but like the stream adapts to the contour of the ground, I need to adapt to what works with those around me.
DevOps is about being better and more efficient. It's a lot of what many of us would aim for without many restrictions if we were motivated. It's also about driving automation for repetitive things. The cloud is about flexibility and moving CapEx to OpEx. That being said, take this with a grain of salt, as I only can see what you wrote.
In terms of helping people work with and analyze data, I'd likely sit with them a bit in my spare time, or ask some async questions to try and understand how I might help. Can I automate something, can I improve quality, can I help them work with data better in an XLS by organizing it. As an example, I worked with someone that loaded data into sheets and then added columns and manipulated numbers. However, reloading data, or adding more sometimes was a pain. I organized this more into a table structure in one sheet and then the analysis elsewhere, so as we changed the table (or loaded more) the analysis wasn't affected, we didn't need to alter formulas.
I might help with that here, then show build a Power BI Desktop that looks at the XLSX and show it to them. A lot of people find the visualization really handy when looking at data, especially with slicers. That gives you a chance to learn and you might help them. If they don't like it, ask for what might help them, or try something else. It's about experimenting (DevOps principle), failing fast, doing something small, getting feedback.
For the cloud, I'd look at a place where my org might be struggling with flexibility. Maybe there's some place I can introduce some automated processing with storage and functions, maybe there's a place where we struggle to upgrade hardware or resize it. Maybe I could find a way that processing data in text files into better analysis (data warehouse) might work better in Synapse, where I don't need ETL. I EL the text files into storage and then let Synapse transform or just read text files and analyze them.
Again, pick something small, something quick you can set up and test and show an idea. Then others might find a reason to adapt your work.
If nothing else, I've had success in the past with simple VBScript (long time ago) or PoSh scripts that do small tasks for people, simple things that they repeat over and over. Building those is a DevOps skill as lots of DevOps automation is working with the command line.
December 21, 2022 at 4:14 pm
My uncle got into programming in the 1980s and as a COBOL programmer was in high demand in the mid to late 1990s due to the Y2K fix. Unfortunately he expected that demand would last forever and refused to learn any new tech. His company tried to send him on courses or to study for certifications - they tried to get him to grow, but he was comfortable in his rut. Eventually they let him go. He never worked in the industry again, and blamed "ageism".
I learned a very important lesson from him. I've been in IT over 25 years now, and I'm always pushing myself to learn new things. You never know which "hot tech" of today could become a standard tomorrow.
There are still COBOL jobs these days, paying well, but few of them. If I were 10 years younger, I'd be tempted to jump there and support someone that still keeps a mainframe going.
I'm like you. Learning more always creates opportunities for me.
December 21, 2022 at 4:26 pm
Problem with learning on one's own [dt]ime is you can really only barely scrape the surface of all these techs going through books and training courses. The ecosystem is complicated enough that to really learn the stuff, you have to get your hands dirty and fail a lot at a real project. That's where it's critical companies invest in meeting those employees with the interest to learn half way.
I agree and disagree. You can't learn everything, but you can pick and choose and in a year, you can get 1-2 skills down well enough to use them in projects. However, it is critical if the company will use something that they invest in people, along with the requirement the people invest themselves at the same time.
December 21, 2022 at 4:43 pm
What I find funny is that all of the too-cool-for-school stuff is what people concentrate on. They don't realize the importance of SQL because they think they know SQL... in most cases, they haven't even scratched the surface. Most of the people I've interviewed in the last decade can't even do the Fizz-Buzz problem with any degree of competence and almost none of them can do a daily total and return zeros for missing days.
I think my slogan for 2023 will be, "Want to learn something NEW? Then learn about all the stuff you don't know about SQL!"
--Jeff Moden
Change is inevitable... Change for the better is not.
December 21, 2022 at 6:00 pm
FWIW, Jeff, it would be great to see a list of common SQL problems that people ought to be able to solve. I'd add in
Be nice to put together a "learn these skills" article with follow ups later on how to solve them.
December 21, 2022 at 8:35 pm
That sounds like a fantastic idea, Steve. Any ideas for a series title, just to add some consistency to such articles? Some of those have already been written though. Like the ones about generating large volumes of data and Crosstabs/Pivots. Should shorter versions of those be included?
On the XML stuff, I'd be a learner rather than a writter. 😀 Maybe we can goad Mark C. into writing some of those. He's a magician in all things that are XML in SQL.
--Jeff Moden
Change is inevitable... Change for the better is not.
December 21, 2022 at 8:38 pm
Not sure about a title, but something like the lines of "1 of the top 10 query problems you should know how to solve".
I'd look to present some scenario, sample data, results. Maybe a test harness. Then ask people to solve this themselves, they ought to get the results. Then maybe a week later we publish a solution or two and explain how to solve the problem. I'd hope that a few people would write setup and solution articles, so it's not just one person.
might not be great for interview tests, but would be neat if people interviewing actually read the articles and learned something.
December 21, 2022 at 8:59 pm
Agreed. Definitely not a one man show on this type of thing.
It's funny that you brought this up because I was thinking about how many things I and others have easily solved using INTEGER MATH. It seems like a lot of even well experienced folks don't consider it. I was looking at some questions about the "Julian Dates" (a mistaken but common name for the problem) that JDE uses because I had just answered one using Integer Division. Only 1 person out of a couple dozen answers I've seen used it. The rest was based on some cumbersome Left/Right code and even a couple of well known heavy hitters managed to actually produce incorrect answers because of all that. They did the same thing with UNIX Timestamps, which is the reason why I wrote that particular article and solved it with Integer Division.
"Top 10 SQL Queries You Should Know" sounds like a great title, especially for SEO purposes, even if we go over 10.
--Jeff Moden
Change is inevitable... Change for the better is not.
December 21, 2022 at 9:06 pm
Thanks for the insight! The article and responses were timely for me as I ponder the proverbial question of "What's next?"
December 21, 2022 at 9:08 pm
Hmmmm.... maybe a kickoff with the most basic and, yet, one of the most important things that isn't actually taught in any beginner classes that pretty much laid a foundation for me in my early days... "How to Count". It would do some comparisons of methods and end with solving the Fizz_Buzz problem with a promise that future articles would teach about how to create series of dates to solve "The Zero Sum Days" problem.
--Jeff Moden
Change is inevitable... Change for the better is not.
December 21, 2022 at 9:12 pm
Maybe a multi-author Stairway to make such articles easy to find as a group? To make it as high quality as possible, perhaps have a couple people do a review on each article and maybe even setup up an overall set of minimum requirements. Like you imply, must be demonstrative, etc?
--Jeff Moden
Change is inevitable... Change for the better is not.
December 21, 2022 at 9:19 pm
Last observation, we've been experiencing a lot of turnovers recently. One of the biggest positions to leave is the head of our Project Management Office. In an all-IT meeting, she said the most damning thing of all (although she backpedaled a little from the comment). She said that after working here for a few years, the issues we face and the unwillingness to change, lead her to decide to move on. She said that the changes required in culture are significant enough that it might require legislative action to fix. (We're a large state department, so this comment, although extreme is possible.)
Rod, having seen you share your frustrating circumstances before, I was half wondering if you need to be a whistleblower ... along the lines of "state agency staffed by technical professionals who refuse to use latest tech." But I wouldn't want you to get fired, or worse.
Trying to figure out the world of SQL as marketing consultant for SQL Solutions Group https://sqlsolutionsgroup.com/
December 21, 2022 at 11:52 pm
Rod at work wrote:Last observation, we've been experiencing a lot of turnovers recently. One of the biggest positions to leave is the head of our Project Management Office. In an all-IT meeting, she said the most damning thing of all (although she backpedaled a little from the comment). She said that after working here for a few years, the issues we face and the unwillingness to change, lead her to decide to move on. She said that the changes required in culture are significant enough that it might require legislative action to fix. (We're a large state department, so this comment, although extreme is possible.)
Rod, having seen you share your frustrating circumstances before, I was half wondering if you need to be a whistleblower ... along the lines of "state agency staffed by technical professionals who refuse to use latest tech." But I wouldn't want you to get fired, or worse.
You also have to be careful that that "latest tech" is actually "better tech". Way too frequently, it's not actually better.
--Jeff Moden
Change is inevitable... Change for the better is not.
December 22, 2022 at 12:58 am
JRuss wrote:Rod at work wrote:Last observation, we've been experiencing a lot of turnovers recently. One of the biggest positions to leave is the head of our Project Management Office. In an all-IT meeting, she said the most damning thing of all (although she backpedaled a little from the comment). She said that after working here for a few years, the issues we face and the unwillingness to change, lead her to decide to move on. She said that the changes required in culture are significant enough that it might require legislative action to fix. (We're a large state department, so this comment, although extreme is possible.)
Rod, having seen you share your frustrating circumstances before, I was half wondering if you need to be a whistleblower ... along the lines of "state agency staffed by technical professionals who refuse to use latest tech." But I wouldn't want you to get fired, or worse.
You also have to be careful that that "latest tech" is actually "better tech". Way to frequently, it's not actually better.
Indeed! Thanks for the assist.
Trying to figure out the world of SQL as marketing consultant for SQL Solutions Group https://sqlsolutionsgroup.com/
December 22, 2022 at 3:30 pm
FWIW, Jeff, it would be great to see a list of common SQL problems that people ought to be able to solve. I'd add in
- running totals
- deserializing XML and/or JSON since people stuff those into databases
- pivoting/unpivoting data
- generate a large volume of data with a tally table
- split strings based on some delimiter that STRING_SPLIT can't handle
Be nice to put together a "learn these skills" article with follow ups later on how to solve them.
I like this suggestion, Steve and Jeff. I certainly would read that series. For example, that running total problem, I've only solved that with different report writing tools. I'd love to see how that can be done in SQL.
Kindest Regards, Rod Connect with me on LinkedIn.
Viewing 15 posts - 16 through 30 (of 32 total)
You must be logged in to reply to this topic. Login to reply