tumba

This is true of some programming, but not all. If I need to do some quick data munging, or build a CRUD application, lisp may not provide any benefits over another language in which I am equally skilled. But if I am programming in a complicated domain or doing exploratory work, I would rather be using lisp than any other language.

François-René Rideau gave a talk called "Using Common Lisp as a Scripting Language" at this year's ILC [3] that covered these subjects in detail, but I don't believe any videos have been posted.

[1] http://cliki.net/cl-launch

[2] http://tunes.org/files/asdf3/asdf3-2014.html

[3] http://ilc2014.iro.umontreal.ca/program.php

Ironically, in large scale (SAP-class) enterprises today, the integration problems are often relate to the fact that they are running multiple ERPs they picked up through acquisition and run customized business rules that cannot be cost effectively moved. Look at large scale master data management tools to get a sense of this type of problem.

Much of my work involves mid-market ERP from Microsoft, Sage, Infor, etc. These vendors can give a smaller company an entire integrated suite of tools, with great support, an ecosystem of ISVs fully implemented for $200-300k, including consulting. That remains a compelling value proposition and less risky than hiring programmers, who these companies have no idea how to manage.

One company to watch is Infor. Their strategy involves traditional ERP (usually products tailored for a vertical) at the core, with ancillary products integrated through standard middleware or APIs and delivered through Amazon cloud services.

Or perhaps, to enable program resurrection. Naur explicitly contends that theory is not something that can be expressed. I disagree. Communication from one mind to another is never perfect, but humanity has thousands of years of experience communicating mental models about the world.

My experience managing Windows servers is rapidly improving. More and more servers are running Core and I'm performing ever more administration remotely by command line using PowerShell.

I bring up Windows Server because, on that platform, Microsoft is clearly articulating a change in their philosophy of system administration. The interactive or scripted command-line is the future of Microsoft server administration. It exposes the underlying management and instrumenting capabilities of Windows in highly useful ways and allows consistent workflows for managing traditional deployments and virtual/cloud systems. They have been clear about this direction for years, and they have facilitated a long transition by continuing to provide the option to use full graphical installs, and by providing management consoles for their major products that simply execute PowerShell commands in the background.

Maybe I just haven't seen it because I don't follow the consumer facing groups at Microsoft very closely, but I wish Microsoft would take a similarly approach for the desktop.

Whenever I read these articles, I am thankful I am not an ISV developing for the Windows desktop.

(1) Insights and Ideas. The reason these are valuable to me is the thoughts they provoke in my head. I try to make most insights I come across "mine" by writing them and the ideas they prompt in a memo to myself or even simply on an index card. Of course, I always record the source, even if I don't record an exact quote, to avoid plagiarism if I use these notes in something published. I also have a computer file of quotes on various subjects. These are usually either humorous or wise rather than topical.

This process is much more time consuming than slurping them into an electronic storage system, but for me, the cost associated with this technique is a feature. The "friction" involved in pondering and writing about ideas forces me to be discriminating and not waste my time thinking about or storing ideas that aren't valuable to me. Perhaps other people don't need this discipline, but it has made me more productive than when I have used tools optimized for rapid storage.

(2) Code. Functions or programs that demonstrate techniques or features go into a well-named file in a directory reserved for each language or product I work with.

(3) Facts and formulas. I don't record many of these, preferring instead to have (or have access to) good references material, but when I do, I either print the source, or write them myself on an index card and store them in a paper file per subject.

[1] http://books.google.com/books?id=A6ZO2D6ayNwC&dq

"Client shall have a perpetual, irrevocable, nontransferable license to use and copy the materials and deliverables created, discovered, invented, developed or prepared in the course of this agreement (“Deliverables”) and prepare derivative works based on the Deliverables for its internal use. All other rights in the Deliverables remain in and/or are assigned to Consultant."

I use a simple sequential filing system for articles. I assign incoming articles a serial number. I have a file that allows me to browse these by subject and author, but the serial number makes them easy to reference when I want them.

[1] http://www.amazon.com/The-Handbook-Financing-Growth-Transact...

http://portal.acm.org/citation.cfm?id=361612

Paul Graham even posted it in its entirety here: http://www.paulgraham.com/knuth.html

This question hinges on how you define the terms science and art. I like Knuth's definitions.

My choice of a diesel was not simply to achieve a lower fuel cost per mile. Maintenance costs and reliability are also better due to the lack of certain systems within a diesel engine (i.e. no spark plugs, alternator, etc.).