Archive for September 2009
Over the course of the last few years I have read quite a lot of business books and blogs. I’ve gleaned a lot of valuable information from these sources, but in the back of my mind there are typically also a lot of bits of information that I fill in or gloss over when reading from these sources. You see, coming from a science background, I am much less likely to take some uncited and unresearched nugget of conventional wisdom at face value. I’m not trying to say that I’m smarter than these guys, I’m just trying to say that I treat much of what is said as hypothesis rather than fact. Often times there will be glaring holes in the story or unanswered questions when reading the accounts of the successful business or entrepreneur being described in the book or blog post.
I don’t know what it is that keeps me coming back to fill up on more business knowledge time and again from these sources, but there must be a deep seated need for intellectual gratification that I expect to find – or maybe it is a play on the desire to succeed or find the missing bit of information that was holding me back. From experience I have found that it is typically inspection, reflection, and introspection that leads to breakthroughs. Nuggets of wisdom are valuable for initiating these cycles of reflection, but on their own don’t typically solve any problems directly.
Over time, however constant skepticism becomes very tiring, as you are always second guessing yourself and what you are reading. I really wish I had an answer for this but I really don’t.
I can’t help but think that the way business books and business education curricula are so based on case studies exposes them to huge amounts of confirmation bias. It works like this: pick your favorite business theory and you are sure to find several instances of companies apparently following the principles to succeed. Of course there are probably many that follow the principles and fail also, but not only are the stories likely buried, but it is also easy to attribute the failure to some other confounding variable.
I realize that a lot of blogs are sensationalistic in order to keep readership high. The bloggers have their ‘people’ who religiously follow every post and comment en masse any time a new post is put up. The bloggers know that to survive they have to keep coming up with nicely digestible soundbites, nuggets of wisdom and top ten lists of why you are doing things wrong.
I’d like to see less sensationalism among the business book writers, gurus and bloggers, but I know that it is never going to happen. For the same reason that consultants keep an air of mystery and omniscience to be able to keep their rates high and their customers calling back, publishers and writers need a similar air to keep their businesses alive.
Maybe the ideal is to read blogs of writers whose primary function is not a business blog writer but a CEO/CTO or other executive. They presumably have the experience to speak on business given their position, but they are not completely invested in making money on their writing. Of course there could be the issue of public reputation and disclosure. The competitive nature of business is such that you don’t want to accidentally give away some strategic elements of your business to your competition right? An executive blogger may also be more guarded in what they will venture to say on a blog also. More radical ideas may not be expressed due to fear of ridicule as a company leader.
I also can’t help but feel that as technologies and markets change, we scramble for answers and guiding lights to help us navigate the changing waters, and we are likely to latch on to theories that are more like memes than business theories. Since markets and technology can change much more rapidly than our education systems and mechanisms for coping with business realities, half-baked explanations of why some business practice is good or bad travels more like an internet meme.
My thoughts on this came out a bit less clear than I had hoped. I may attempt to rewrite this post or write a follow up.
I’ve recently been going through a lot of past audio work and reviewing my approaches/techniques. I have a lot more ideas, but I’ve been away from it for a long time. Hopefully I can create some new stuff, but in the meantime I’m cleaning up several synths for release that think would be useful. Here is a pretty advanced subtractive synth that I did that was based on the emagic ES2 design, but with a pretty impressive modulation matrix.
Overall structure of the synth:
Detail of the modulation matrix:
EcStatic is a subtractive synthesizer whose design was inspired in part by the ES2 synth from emagic. This synth has a similar 3 asymmetric oscillator design, but with an extended routing matrix and effects processor.
copyright 2003 tekrosys
Recently I’ve using my old tablet PC for sketching out UI designs, and I am constantly reminded how bad the handwriting recognition is on Windows XP Tablet Edition. There are several input methods ranging from the annoying ‘write between the marks’ monospace and a more natural freehand method. However, the system constantly mistakes letters for numbers and vice versa, and using the freehand system, it is much more difficult to correct mistakes. This gets to be so annoying that I end up using the onscreen keyboard and just use the stylus to tap out what I want to enter. I found myself thinking ‘it would be great if I could somehow tell it that I am writing a number instead of a letter’. But wait, I had a device oh, six or seven years ago that had no problems with this. The Palm Pilot.
So the other day, I dug out my old Sony Clie. The Clie is a PlamOS device, featuring the same Graffiti handwriting recognition system that all Palm devices used to have. There is a dedicated space at the bottom of the touch area for handwriting input. The space is divided into two sections, one for letters and one for numbers. If you wanted to make a capital letter, you wrote in the center, crossing into both halves. For the small concession of having to be explicit about what you wanted it to do, you were rewarded with a very small error rate. However, this is all notably absent from Palm’s later devices – the Treo and the Pre.
Taking these two things, namely the inadequacy of handwriting recognition in Windows Tablet Edition and the lack of Palm’s inclusion of any sort of handwriting recognition in their devices, one wonders where handwriting recognition as an input method is going.
Since mobile devices are increasingly doing away with the stylus, it seems unlikely that things will move toward handwriting recognition again. Furthermore, if tablet devices adopt a touch interface rather than a stylus interface, I don’t see how handwriting recognition will survive on the platform. Even if the stylus remains, users may prefer using the touch interface primarily, making switching to the stylus a chore.
I think that handwriting recognition as it existed in the days of Graffiti and even as it exists now on tablet PCs is dead/fading, if for no other reason than I see the stylus losing the war of the input devices to touch screens. The stylus is unwieldy and inconvenient at best (how do you manage a tablet pc with one hand and the stylus in the other?), and dorky at worst (ever see a guy in a bar scribbling away at their Treo with the stylus?). Furthermore, I see innovations on the software side such as Swype picking up where Graffiti left off, offering fast and intuitive text input for touch devices.
I’ve had some thoughts on writing recently. As you may or may not know, as a founder of Ubernote, writing notes is a habit that I am very serious about, and is near and dear to my heart. However as critical as notes are, I think that writing clear and finished pieces of work is also very important.
That much is not exactly earth-shattering.
What I want to propose though, is that it doesn’t count unless it is public. Steve Yegge’s post, You Should Write Blogs comes to mind to back this up.
This is a disconnect that is a fundamental problem for me I think, and one that I am working on in gradual steps through a combination of habits. I could be wrong about what the most effective set of habits is, but here is what I’m starting with:
Habit #1 : Posterous Blog
I have started a Posterous blog as a place to write very small or personal narratives that I don’t want to put on my main blog. This has turned out to be the single biggest breakthrough for me. Since I don’t have a set of followers like I do on Twitter, I can sort of dump stuff here in great volumes without feeling constrained by the feeling that I am spamming stuff out at an annoying rate for anyone. The second advantage is that I am more at ease to write something shorter or more informal here which also turns out to be a huge help. Many times I will start here and end up with something that I will want to expand on to post on my regular WordPress blog.
Habit #2: Twitter:
I listed this one second, since I think that while this habit had some real initial benefits for me, it has now been relegated to building relationships. Writing tweets does help you with good post titles and headlines though! The 140 character limit means that you are forced to really convey the message succinctly, which is perfect for coming up with post titles.
The advantage that Twitter had for me was that I could write many more posts than I could with my blog, since they were so short. I think that Posterous is a better solution for small posts now though, since it encourages real writing, even if it is only a single paragraph. Writing is the art of the paragraph, and twitter will not help there.
Habit #3: WordPress Blog
This is the heaviest habit that I have for writing at the moment. I could see that maybe I would move up into the world of essays after having gotten more comfortable with the blog format. The Blog is actually the first of the three habits here that I started. I have found that my blogging has changed a bit since the introduction of Posterous especially. With Posterous now absorbing smaller posts and thoughts, my WordPress blog is now the place for bigger ideas or posts. However, many times a post ends up taking much longer than I had anticipated, and so I have nearly 100 posts that were never posted! This is not great, since my whole thesis here is that the writing must be public in order to ‘count’
The writing habit is an important one, I think, since the act of writing is a deliberate one, and as such must be approached with deliberate practice with the goal of reducing the mechanical aspects to rote and allowing one to focus on the message. I’ll continually modify these three habits, or perhaps add/remove some to further this goal.
I have been thinking about writing a few posts about learning techniques and methods. I don’t want to turn my blog into a productivity or life hacking blog, but there are a few thoughts about the way that I approach learning that I wanted to write down somewhere and get out.
When I think about approaching a totally ‘green field’ area of knowledge for me, I do probably the same thing most web-savvy people do: I start hitting Google. However, this is where I think there is a divergence in behaviour between different people. The exact search terms use in a search are an art form unto itself, and interpreting the initial results and iteration/refinement of the search all play critical roles in how the initial material looks.
I envision my search starting off on the ‘shores’ of my own knowledge. I will often think about what is the closest known peninsula of knowledge that seems to jut out in the direction of where I think I need to go and start from there. After that, searching and gathering new knowledge is a lot like little islands popping up on the horizon, and it is my job to build bridges between the relevant points of knowledge.
Ultimately though, the bridges over the sea of new knowledge must end up on land again in order to ‘close the loop’ of new knowledge! I think that this is one of the most critical aspects of retaining the knowledge and forming a foundation for expanding on the newly learned material. It also forms the basis for true understanding. Once the loop is complete the bridges can give way to highways as the information becomes integrated into the corpus of well traveled pathways in my brain. Eventually, the areas between the loops can be filled in and even becomes solid dry land of knowledge, and forms a new jumping off point for the process to begin anew.
Here is probably the first full synthesizer patch that I ever made. I was impressed with the Subtractor synth that was included with Propellerheads’ Reason suite, and I wanted to have a standalone instrument version of it. I had the idea of making my own version of it using some sampled single-cycle waveforms from the original synth. I happened to be using Native Instruments’ Reaktor then, so unfortunately you will have to have it installed to run this patch. It is a commercial piece of software. This synth turned out pretty well, so maybe I’ll convert it to something else that is open-source. Note that I sampled waveforms directly from Reason for the oscillators of this synth to try to preserve the character of the original.
I originally did this using Reaktor 3, and I converted it and re-saved it in version 5 to post here.
Here is a screenshot of the panel layout:
And here is a screenshot of the dsp structure:
The following is the ‘about’ info text from the patch:
an emulation of the propellerheads’ subtractor synth from reason. only the first 4 waveforms are included though, as i didn’t want to sample all 32. to make up for it, you can mod the waveforms with any other waveform instead of being limited to modding the wave with itself. also the lfos and envelopes may be assigned to more than one destination. every other feature has been emulated with the exception of legato mode. enjoy! [tekrosys]
I was digging through old music production stuff, and I found some patches that I did using Pure Data. Pure Data is a graphical programming language for music performance and sound data manipulation. At the time that I wrote this, I really needed a live performance sequencer that had extensive keyboard shortcut coverage, and that could run on a low powered laptop that I had. While this patch met those requirements, the MIDI timing of PD on that old laptop turned out to be too inaccurate to be of much use as a sequencer. Maybe I had something wrong configuration-wise. I’m releasing this in the hopes that someone will find it useful.
Below is a copy of the online help file that can be accessed by clicking on the ‘pd help’ icon at the bottom of the sequencer UI.
peeq is a simple pattern sequencer for the pd environment. it started out as a simple grid to bang out midi data, but has evolved to support multiple patterns and a small playlist. I wanted an extremely fast way to output midi data in a live situation, so flipping between patterns and entering row data can be done very quickly using the keyboard. the design is now modular, so rows may be added by simply creating instances of the row objects. The number of patterns may be increased by changing the limits of the arrays used to store them.
space bar – starts/stops sequencer
+/- – increment/decrement pattern
c – clears current pattern
</> – increment/decrement row selection
[/] – increment/decrement channel number of selected row
;/’ – increment/decrement note number of selected row
l – toggles playlist mode on or off
qwertyui and asdfghjk – these keys activate the steps of the currently selected row
chase – enabling this forces the pattern editor to follow the playlist.
1-4 – number keys assignable to pattern macros. actually any key may be assigned and any number of macros can be defined. look in the ‘keycommands’ subpatch under ‘guts’.
copyright:tekrosys.2003 (dan newcome)
Note that PD must be installed on your system in order to run this patch.