kusmi

We can agree that everything that needs to be pinched and squeezed can be pinched and squeezed with Latex. So, creating stylish modern looking tables, full page width graphics overlayed with expensive commercial fonts, custom vector graphics and diagrams, is feasible. You would never even recognize it was done with latex. This makes sense at corporate scale painting reports, white papers, invoices, in company colors, logos, and styles. Churning hundreds of styled docs. That's the use case I see for Latex, but no one goes around showing these off, I assume because making styles like this is a job, and the styles themselves assets.

I like bash for the same reason I like emacs, in that no matter what the environment is like, I can usually count on my bash scripts to work. I keep them in emacs org-mode files where I store them in src code blocks. I can tangle multiple code blocks into single executable scripts to different directories. Check out org-mode babel, tangeling, and noweb. Keeping all my bash code in a single file solves my issue with having to dig for that one script I wrote that one time because I forgot how to do this one thing ...

If you aren't running Linux on your desktop yet, consider it. Full immersion is a fast way to learn.

Currently reconstructing the entire thing to production spec, as an AWS AMI, perhaps later polished into a personal knowledge base saas where the cleaned and sorted content is public accessible with REST/cmis api.

This project has single handedly eaten almost a third of my life.

There is too much information to stay on the bleeding-edge of things without serious commitment of resources, which start-ups don't have. My intention was to track the content a group of people go through in a day and visualize connections in the data that may not have been obvious before.

Essentially, it's meant for harvesting IP in biotech field.

*Forgot to mention the best part: Backend pools these full-text documents, cleans and parses for NLP, then generates meaningful tags, and organizes documents in an auto generated folder hierarchy which is based on word2vec/doc2vec and content clusters. Whole thing runs on a dedicated server with two 1070 GTX video cards for the NLP work which is training and re-evaluating constantly as new content pours in.

Altogether it was 2-3 years of work.

It is more commonly referred to as the prisoner's dilemma.

This is a strange article, and I'm curious who the author wrote it for. It's certainly not written for biologists familiar with the research she refers to.

If not, perhaps I can interest you in a demo of one? If you don't see an increase in turnover, well make the trail period free!

Coming out of university with a BS, graduates today are far from well-equipped to perform in corporate upper middle-class jobs (the jobs promising upward mobility). I think this has most to do with an increase in the amount of data companies deal with. White-collar employees need to know not only the ins-and-outs of their industry, but they must also know how to use increasingly complicated enterprise software, and possess more advanced analytical skills to operate with more complicated data. As a result, graduates with a 4year degree are stuck for 5+ years in internship limbo. Companies will bleed money training interns who literally can't offer anything of value and instead will use up the work hours of the skilled employees who train them. Hence, <30k salaries.

I think this is why there is a trend for the more math-heavy STEM fields to flee the coop earlier --they spend less time on the software and analytics training to become valuable sooner.

Certain degrees like health, or design are unlike general STEM fields in that they train students for specific technical jobs rather than teaching broader critical thinking/problem solving. So, they get jobs right out of school (5year engineering programs...?).

PhD's are in the best position to flee the coop quickly because they are paid a livable wage (in most cases) while in graduate school. It's also not terribly difficult to get into a PhD program (completing it is another story). After receiving their doctorate, PhDs rely on tight-knit academic networking within their field of study to do their post-doc. However, this is where many of them get stuck, receiving a salary <70k (often far <). Increasing competition for faculty positions has pushed many a post-doc out of academics (though, by this time I feel they can compete well in the corporate sector).

What I can't explain is law students. I thought there was a bubble leaving most of them jobless. I presume that maybe a law degree is relatively more helpful in scaling the world of big-data (and 'big-regulation') making them prime pickings for corporate jobs earlier than the rest.