Forum Posts Feed
Processes not tools
< Next Topic | Back to topic list | Previous Topic >
Posted by J J Weimer
Sep 25, 2018 at 01:12 PM
Respectfully, I think some of the depth of your message may get lost in the abruptness of your opening statements. Let me see if I can dig deeper.
1. When I am truly focused on the process, I do not care about the _source_ of the tool, I care about whether it is suitable to do the job. You seem to accept a platform agnostic approach yet also seem to fuss loudly against single-developer software. What I think that you are saying is that you avoid software that has a locked-in or dead-end approach in its design. I agree with this philosophy.
2. Well. As an abrupt counter ... I guess that means we should drop our use of paper and pencil or clay tablet and stylus or cave wall and chalk. But then you turn around and praise tools that you _do_ use. What I think you are saying is that you avoid using tools for their own sake. I agree with this philosophy.
3. You are saying “Internet” to mean only blogs, forums, and podcasts. The internet has a wealth of other resources that are worthy of solid praise.
4. You embrace it, but then you define caveats to your “embrace”. This sounds more like a “I date anyone and everyone ... but they must have blue eyes” kind of statement. What I think that you are saying is that you use the cloud to the best of its potential in your work. I agree with this philosophy.
5. We all can stand a reminder to continue our good habits for life-long learning.
Posted by MadaboutDana
Sep 25, 2018 at 01:32 PM
I think there are some great points here. And if I weren’t an inveterate CRIMPer, I would embrace your philosophy wholeheartedly!
On the other hand, the term “mainstream” is a problematic one (as a few forum members have already pointed out). As a lurker, you’ve no doubt spotted that many of us incline - possibly increasingly? I haven’t done an objective analysis of forum trends, but anecdotally, it seems to me that a certain trend is apparent - to apps that hold data in multi-platform or totally open formats such as plain text (in which I include various flavours of Markdown) or HTML.
These seem to me to represent a good definition of “mainstream”, and perhaps a better one than talking about mainstream apps. Using long-lived, mainstream formats like .txt and .htm(l) is a good way to avoid the lock-in you so accurately describe (ah, I have a nice collection of gadgets, just as you do: all I can do nowadays is get them out, gaze fondly but sadly upon them - my old Psion 5mx’s (I blush to admit I have 2, including the Ericsson variant); my AlphaSmart Dana; my Compaq and Toshiba WinCE devices; my LG Phenom, etc. etc. etc. - and then put them away again). Sad really. But jolly good fun at the time.
Of course many of those devices are theoretically still usable, precisely because they ALL support text and HTML (although things like UTF8 support etc. can make even these basics slightly less than totally predictable). And this, I suggest, is where yer average CRIMPer has room for a certain amount of manoeuvre. I love experimenting with tools that use these basic formats well - my recent discovery of Zettlr being a case in point. Which brings me to your second, interesting point:
It’s best to spend time familiarising yourself with EVERYTHING your “mainstream” tools can do, rather than wasting said time playing around with other apps.
Really can’t argue with that. You’ve got me bang to rights, guv’nor. Knowing your mainstream tools in depth is, of course, one of the secrets of making best use of them. But here’s a thought:
The ongoing development of mainstream tools can tend to the sluggish as complacent developers cease to be quite as inventive, quite as imaginative, as they were originally, precisely because they KNOW their tools have made it into the mainstream, and are therefore irreplaceably established in the marketplace. Hence the seductive pleasure of CRIMPing: new, small, independent developers are often so exciting! Have such amazing new ideas! Do precisely the things you wish the Big Boys had done years ago! Think up totally new ways of handling/analysing/manipulating information! Look at SheetPlanner, for example - full of exciting potential! Not doing anything particularly novel on one level, but bringing together a whole raft of features in a way nobody else has done to date. Watching apps like this unexpectedly succeed in the marketplace, in the process challenging or even displacing the “mainstream” apps one tends to use either from force of habit, or because one has no option (e.g. Microsoft Office, Adobe Suite), is an uplifting experience. Look at the amazing success of Affinity, for instance - displacing Adobe’s well-established empire! Wow!
So while I respect your philosophy, I enjoy experimenting with data management approaches far too much to forego the pleasures of CRIMPing. But thank you very much for sharing!
Posted by tightbeam
Sep 25, 2018 at 02:01 PM
I did read your whole post, and while I also like to “fool around” with the software - much of it very good software - that independent developers create, I would not “rely” upon that software, as the chances of it being abandoned are quite a bit higher than Microsoft suddenly throwing in the towel. I believe an annual subscription to Office 365 costs $70/year, or $5.84 per month. If that cost is too onerous, then it’s time to put up a lemonade stand and earn a few extra bucks.
“Small developers work to satisfy the needs of the customer.”
Well, no. I imagine most customers come to small developers because they see that the developer already has created software suitable for their needs. It’s true that small developers give the appearance of being responsive to customer input - mostly because they have few enough customers that it’s possible to do so - but typically their pace of development is glacial, lacking as they do sufficient time, money, and infrastructure. How long, for example, has the new version of MyInfo been under development? Even Workflowy, with its largish customer base, went through an Ice Age of sorts recently, woken not by their grand concern for the needs of customers but rather the encroachment of Dynalist.
You probably meant to say, “I’m not criticizing that choice, but I would be uncomfortable relying on Microsoft[, exclusively], and with that I agree. I don’t think anyone can successfully argue, however, that Microsoft (or Apple or Google) isn’t *reliable* in terms of its ability to stay in the market.
Stephen Zeoli wrote:
I didn’t say I don’t use Microsoft products. I am forced to. I am saying
>I don’t want to rely on them exclusively. Microsoft’s aim is clearly to
>herd everyone into their Office 365 annual subscription system. When a
>company is that big, it usually means the customers have to adapt to the
>changes based on the company’s needs, and not the other way around.
>Small developers work to satisfy the needs of the customer.
>
>I don’t use Apple’s Pages or Numbers. You’ve got to trust somebody’s
>hardware and operating system.
>
>If you read my whole post you’d see that I addressed small developers
>and Scrivener.
>
>tightbeam wrote:
>Stephen Zeoli wrote:
>>>I’m not criticizing that choice, but I would be uncomfortable relying
>>on
>>>Microsoft.
>>
>>How come? Are you also uncomfortable relying on Apple? the Scrivener
>>folks? any home-brew, cottage developer?
>>
>>
>>
Posted by Amontillado
Sep 25, 2018 at 02:25 PM
Pixelpunker wrote:
>> Amontillado: On the other hand, even papyrus and dry desert caves
>> aren’t perfect.
>
>Sure, but the timeframe of this digital stuff is a joke. I can’t even
>read 10-year-old files.
>
Nor can one always read 10 year old newsprint. It’s not an archival medium. Neither are old word processor files good candidates for forever.
On the other hand, 40 year old files are perfectly readable with anything, if they are plain text files.
tightbeam wrote:
Stephen Zeoli wrote:
>>I’m not criticizing that choice, but I would be uncomfortable relying
>on
>>Microsoft.
>
>How come? Are you also uncomfortable relying on Apple? the Scrivener
>folks? any home-brew, cottage developer?
>
>
Apple Pages’ native files are in a proprietary, closed format. That detracts from the value of Pages, but who cares? You can export to plain text.
Scrivener stores everything you write in individual files, all either plain text or RTF, and you can export to plain text. Or epub, that’s just plain text, too. Epub files, like docx files, are just zip archives. Unzip an epub and you’ll find xhtml or html files carrying the payload (plain text), and CSS files carrying the formatting.
Ulysses stores plain text files, but they are a little hard to find. That would bug me, but it’s ok. It will export to plain text and other formats.
There are trivial differences in plain text “format” files. There’s not really a format, it’s just a series of characters, but there is inconsistency if you look really close.
Unix set a standard for text files where line endings were done with a linefeed character, called a newline in Unix circles. Microsoft decided on carriage return/line feed pairs at the ends of lines, and a (unenforced) control-Z at the end of files. Apple went, initially, with just carriage returns for line endings. Apple has since mended its ways.
That’s inconsistent, but hardly a big deal. The Unix (or Mac) tr command (transliterate) will convert files whichever way you want.
More important, though, is to keep refreshing your archive media. Not only would Apple II diskettes likely be long gone from deterioration and constant magnetic influence, it would be a chore to find a working Apple II to read them on.
We know plain text ASCII will be around for a long, long time, because since the dawn of BNF all program source code has been plain text ASCII. Vi and emacs are not going away, whether they work well for you or not.
So, can you read 10 year old files? Of course. Your mileage varies according to how well you supported your future self, 10 years ago. Or 40 years ago.
As far as changing software like undershirts, I’m somewhat guilty. When I find something that really works for me, I stick with it. I wouldn’t have de-installed Ulysses, but I got into a snit over subscriptions. For me, Ulysses works so well I made my peace with their subscription model.
Posted by Pixelpunker
Sep 25, 2018 at 07:13 PM
> Stephen Zeoli: I would like to know your definition of “mainstream.”
> It seems to me, on a PC, that means you’re using Microsoft products
> exclusively. If not, then your definition of mainstream is looser than
> mine.
I am thinking about the usual, Office, Sharepoint, Exchange, but also
Lotus/IBM Notes would fit the bill.
> J J Weimer: Respectfully, I think some of the depth of your message
> may get lost in the abruptness of your opening statements.
I certainly don’t want to rile anyone up over their use of finely
crafted software tools made by passionate developers. It is great news
for example that after years of monopoly blandness in the last ten years
serious alternatives to Microsoft Word have sprung up. I also get that
with all the customizing and adaptation of one’s personal workflow one
may get deeply attached to a particular program or platform.
So let me postpone the discussion about the merits of indie programs vs
mainstream programs and talk about the **process** of personal
information management again:
First, what is my objective?
I am talking specifically about personal information management and a
finely curated collection of information rather than a large database
that can only be mined with computer tools.
What I have right now is a hard drive that degenerated into some sort of
black hole. I only collect and rely on search to find anything. I
postpone organizing, pruning and processing the information I’ve
amassed. I also won’t just dump it all because time and time again if
found nuggets of gold in there.
Did you know that Michael Jackson had a full-time archivist that would
categorize, label and archive everything he wrote or recorded? Since I
don’t have the funds for a personal librarian would a tool help that
could organize the information for me?
And even if that tool could at least categorize my information would
that solve the problem of assessing the personal value of each piece of
information? Would it prompt me if the information is time-sensitive?
What method of organization will I use? Put it in a traditional folder
hierarchy? Use aliases so that items can live under multiple headings?
How do I name the headings? Do I tag items (there’s the problem of
synonyms or inconsistent use of tags). Do I use one- or two-way
hyperlinks to make it into a web of associative connections?
Whittaker and Bergman tell me that, surprisingly in light of all the
advanced approaches, the traditional folder or outliner hierarchy is
best and they have some empirical data to back that up. The explanation
for this is that a fixed folder hierarchy leads to retrieval by
locational cues and does not tax the verbal system. It’s sort of like
why the method of loci works as a mnemonic aid. They also determined the
optimal number of items per folder and hierarchy depth that would
optimize the search time by using linear regression:
> “The model suggests an optimization for the trade-off between folder
> size and depth. ... Each additional folder step increases retrieval
> time by 2.235 seconds and each new information item added to a folder
> increases retrieval time by 0.106 seconds. The trade-off between depth
> and size is therefore 2.235/0.106 = 21,0849 21.09 ... We can recommend
> that users avoid storing more than twenty-one information items per
> folder, they should create an additional level of subfolders
> instead.”
As far as naming folders goes everything that helps you remember is
enough, no taxonomy required. This is very different to a public
information space that is shared by many users.
So I will still have to do the digging manually, including trying to
open and convert old file formats. (I noticed that I then loose the
original timestamps, but I want the date preserved I last edited a
document, not the date I converted it to a new format.)
I once stumbled upon in a German book about time management from 1983 on
a checklist for perfect time management. This is Point 4 (my
translation):
————————————————————————————————————-
| I have a system of notes that gives me at any time an overview of all
| my thoughts, plans, commitments and tasks.
————————————————————————————————————-
That is exactly what I want to achieve.
Next up in my reading list: “The Organized Mind”