4 ms·

This is an awesome detailed response from an informed perspective, thank you! I’ll check out the article. Do you think that despite the high error resulting fr

by hsikka 8y ago

This is an awesome detailed response from an informed perspective, thank you! I’ll check out the article. Do you think that despite the high error resulting from the sheer complexity of those interactions, design tools would be useful to researchers? Or is it mostly just long shot proof of concepts with no practical purpose right now?

jfarlow 8y ago
Design tools are absolutely useful to researchers. However, the economics of building/selling software to researchers don't work. The state of the art DNA design app that is the gold-standard for every molecular biologist is not much changed over the past 15 years [1]. And it's free. [1] http://jorgensen.biology.utah.edu/wayned/ape/ http://jorgensen.biology.utah.edu/wayned/ape/
- iso1337 8y ago
  One could argue that the state of the art is now benchling. But I have similar thoughts on the economics of building software to researchers. The domain is complex, the customers require a high level of correctness, and features are often implemented to try things out in the lab and then abandoned once it's a dead-end in the real world. Oh, and most academics don't have money to pay for software, nor do they respect the software engineering process.
  - asdff 8y ago
    I'm in cell biology, and many of the programs used in research these days are run from the command line and are open source and freely available. I don't even have a desktop powerful enough to work with genomic sequencing data, I need to ssh to a supercomputer. There's a lot of documentation so it's really not hard to go from understanding Unix command line syntax to writing short scripts then full on python or Perl with biological examples every step of the way. There are also a lot of very powerful web tools, such as the ucsc genome browser, which are freely available to anyone. Commercial gui-based software is usually sold along with proprietary equipment like a microscope or something, and universally loathed even by people who managed to wrap their heads around it. It's much easier to work with different software from different people or groups if it's all the same syntax, rather than relearning clunky proprietary gui after clunky proprietary gui.

by hsikka 8y ago

jfarlow 8y ago

Design tools are absolutely useful to researchers. However, the economics of building/selling software to researchers don't work. The state of the art DNA design app that is the gold-standard for every molecular biologist is not much changed over the past 15 years [1]. And it's free. [1] http://jorgensen.biology.utah.edu/wayned/ape/ http://jorgensen.biology.utah.edu/wayned/ape/

iso1337 8y ago
One could argue that the state of the art is now benchling. But I have similar thoughts on the economics of building software to researchers. The domain is complex, the customers require a high level of correctness, and features are often implemented to try things out in the lab and then abandoned once it's a dead-end in the real world. Oh, and most academics don't have money to pay for software, nor do they respect the software engineering process.
- asdff 8y ago
  I'm in cell biology, and many of the programs used in research these days are run from the command line and are open source and freely available. I don't even have a desktop powerful enough to work with genomic sequencing data, I need to ssh to a supercomputer. There's a lot of documentation so it's really not hard to go from understanding Unix command line syntax to writing short scripts then full on python or Perl with biological examples every step of the way. There are also a lot of very powerful web tools, such as the ucsc genome browser, which are freely available to anyone. Commercial gui-based software is usually sold along with proprietary equipment like a microscope or something, and universally loathed even by people who managed to wrap their heads around it. It's much easier to work with different software from different people or groups if it's all the same syntax, rather than relearning clunky proprietary gui after clunky proprietary gui.