Making pretty data is my specialty. You got a data set you want graphed up? I'd be happy to help if you'll give me a testimonial and let me add it to my profile 🥰
Discussion
I wish 😂😭 I have.. various directions that I need to pursue more aggressively to actually aggregate my data.
And somewhere in old notes are many of those directions.. and some data sets..
But I'm curious what type of visualization you have worked with! What can you show me?
I basically am working toward the social media front of visualizing networks and users.. so on nostr I am familiar with everything in that regard.. but it's like an experimental dataset.. most of my hours are instead spent on architecture of clients or network instead of visualizing any data, because my data mostly doesn't exist in the form I would deem optimal..
But so on the note of my other data sets, I am positioning myself and aggregating slowly.. but curious what you do 😊
So you can see two publications I did as an economist for the APAC team at IHS Markit (now S&P Global Insights) and my recent publication (a first for me as lead researcher!) at my website kamresearch.global
The honest truth is ideal data doesn't exist. It's always messy and subpar, which is why we just make a note of how future research can improve on our results. Your data is probably much more workable than you're realizing.
Perhaps we can boost each other up! I don't know much about networks, but theoretically it wouldn't be that much different than other data visualization. Let me know what you think! I'm launching myself as independent this year so any projects would be good if only for a moral boost 😅
Ok when I clicked the link I provided it didn't do anything good. So here's another link hopefully will work: https://www.kamresearch.global/
Okay! Interesting! Now I get it. I've been missing the PDF link despite it being clearly at the bottom of the articles. I let skimming get the best of me sometimes 😭
This is really cool! I'll have to dig through these to see what they're about.
This is exactly what I was hoping for basically.
Respectfully, "do you always end up with a pdf?" as your result format?
How do you get from R to pdf?
Yes unfortunately PDF is the leading format for a fully compiled article. R is fantastic with graphics or visuals, then the article is written via whatever org preferences. They are combined, and in the case of formal research orgs it'll also go through a styling. So that'll format everything from the text to the coloring of the graphics. The PDF is typically the final draft output, just because it's less changeable.
I will say, even though it is not official, it's kind of part of the work that most overlook it 😅 When done in it's polished form, it intimidates folks and that's by design. There are ways to communicate these insights to broader audiences but there is no profit in that. But I'm open to making things more open... 🤑
These are nuts. You're really good 😂
I'll have to play with R and get an idea of what it's like. We should chat more soon. Basically I see all of your data here and imagine it being freed from PDF format. But I understand the nature of the universal PDF. Different situations call for different measures. I really connected with what you said about never having perfect data. Curious what we could create. In a bit of a trance lately.
What a fantastic comment! Thank you so much! I am very good at what I do, but I'm also aware of how little what I do means to most folks.
If I could make the insights I try to communicate resonate outside of a traditional PDF, then I do not exaggerate in that I would die happy 🤣
Let's collaborate! Our data is never perfect but it's always fantastic for taking a step towards furthering human understanding!
If you happen to know of some basic data sets that might emulate what you've already produced in any extent, that might be helpful for me to work out the foundation in advance. If the data you've already published is available for me to toy with, that would work. I'm not sure where to get "some random data".. 😂 I believe I can reproduce all of these charts in any flattering or desirable way.
The hard work part.. the impactful / intimidating part.. the profitable part.. we'll have to set that aside and stay technical at first. But that is all why I am here.
Ultimately you can spend any amount of time or money refining data. So of course scope is just as relevant in your line of work as is in mine. Referring to development. Technical debt in my case equates to.. R&D debt.. for you.. is there a term for too much analysis?
Yes, I can send you the data for the GFOA article from this year. Do you have signal? DM me and we can exchange contact info. I could probably recreate the data from the IHS articles, but I don't have that readily available.
Typically statistical analysis is pretty standardized and I'm almost always working on a tight deadline, so there is not too much time burn opportunity 🤣 For the GFOA article I wasted a lot of time trying to develop a language model but I failed. I had been thinking I could "vibe code" something because people boast about how good and easy it is but that was not my experience of it. But I only did that because it wasn't for a client the whole article was an independent flex 💪
The most time consuming part of the work I produce is the big picture insights but that's a result of me being a big nerd who is semi-constantly consuming information related to my projects. I don't bill for that part 🤣