this juxtaposition of words is incomprehensible to me
_ is one exact value: the limit of any sequence growing unboundedly
this one exact value is equal to itself no matter the sequence
adding sequences diverging towards _ and __ doesn’t lead to a single
defined result → NaN, same goes for
, and they should perhaps compare equal, but
they are also clearly distinct in some sense.
-E
On Wed, 23 Feb 2022, Hauke Rehr wrote:
this juxtaposition of words is incomprehensible to me
_ is one exact value: the limit of any sequence growing unboundedly
this one exact value is equal to itself
There is some value in convenience, and this question
didn’t come up for the first time. It’s not to be expected
to be the last time either unless an easy solution is provided.
Taking the “white pixels” approach, there could be something like
pd 'crop xmin xmax ymin ymax' NB. or 'borders' in
How about
foo0`foo1`foo2`foo3 `:0 ''
?
Am 02.04.22 um 19:42 schrieb Devon McCormick:
Hi,
I have a group of monadic verbs for which I want the concatenated result,
e.g.
foo0=: 3 : '0'
foo1=: 3 : '1'
foo2=: 3 : '2'
foo3=: 3 : '3'
allfoo=: foo0,foo1,foo2,foo3
allfoo ''
0 1 2
Thanks for the enlightening insights into some detail
of the workings of JE internal address management.
Much appreciated.
Am 05.04.22 um 04:29 schrieb Henry Rich:
I should have said 'the J symbol AREA is a contiguous array...' The
symbol tables are separate structures (closed-address
Sure.
Most J programmers can’t tell nouns and adjectives apart.
We don’t even have a type function returning grammar terms
indicating what part of speech our entities are. Poor us!
We should have been “more careful.”
Am 22.03.22 um 01:00 schrieb Raul Miller:
Be careful here -- "implementation"
do they? looks all good to me
Am 21.03.22 um 22:47 schrieb Elijah Stone:
(A side note: does anyone know why when I post REPL session
transcriptions, they become misaligned?)
--
--
mail written using NEO
neo-layout.org
Personally, I’m fine with the little syntax highlighting I get in vim.
Much could be improved but I don’t feel I need it.
But there doesn’t seem to be a pygments lexer for J.
I find an APL lexer, though [1]
Has anyone tried to provide one already?
@Michal: could you talk about the way you did
You already understood that some kinds of brackets
don’t come in balanced pairs. But parentheses () nearly
always do, the exceptions being inside of strings,
obviously, and closing ) on a line of its own
(which doesn’t occur here, so I won’t explain that).
(oh, and {{…}} introduced another
-study to inform the design of a serviceable dictionary feature
for J.
On Tue, 1 Feb 2022 at 19:49, Hauke Rehr wrote:
Personally, I’m fine with the little syntax highlighting I get in vim.
Much could be improved but I don’t feel I need it.
But there doesn’t seem to be a pygments lexer for J.
I
another approach that could be helpful
with respect to that question:
try removing that , and rearranging
)@{:)^:y
to become
)^:y)@{:
in the definition
what does it compute now?
Sierpinski =: {{ (((72#:~8#2){~3#.\0,],0:)^:y)@{: ,: 1 y} (>:+:y) $ 0 }}
' #'{~ Sierpinski"0 i.>:15
This gives
So there was some kind of misunderstanding.
Perceived point of view:
There should be a total ordering on the (in theory)
infinite set of J values. Fine.
And maybe more than one (Elijah already pointed at <. vs /:)
in different contexts.
This again drives us to the question:
Do we actually want
I wonder why there’s still so much traffic on this thread
(okay, I’m to blame for quite some of it)
I thought xash published the best correct solution.
Am I wrong?
Am 27.01.22 um 18:42 schrieb 'Mike Day' via Programming:
Pawel Jakubas has moved on to something else in his correspondence,
I’d rather have the shape with respect to the topmost other kind.
First axis numeric:
the numeric shape below which the first symbolic one occurs
first axis symbolic:
vice versa
you may need to check the return value of the adjusted implementation
of the datatype primitive for those “other kind”
I also long time wanted to try it.
I had no urgent need of it, so time just went by …
boxed list of columns if we want mixed datatypes which
would be necessary for a drop-in replacement of dataframes
the list of column names imo should not be separated
but those are implementation concerns
Same here. Let J be free to arrange at will.
Conceptually having an order on the domain
means talking about yet another kind of ds.
(in the integral case, that’s good old arrays)
Am 01.02.22 um 08:55 schrieb Elijah Stone:
I still do not think of maps as having order.
--
--
Where “null” is specified as _. right?
Am 08.02.22 um 05:27 schrieb 'Pascal Jasmin' via Programming:
set function is a combination of JP's set or delete functionality with k/q's upsert, with
"merge" oriented optimizations for bulk oriented upserts. To get JP's upsert
or delete functionality,
To be honest, I’m a bit confused by the stated difficulty.
I haven’t worked much with that value (yet) but this is
what I thought it was meant to/would want it to work like:
I got to know _. as a value that may be present in data,
usually not first occurring as the result of a computation
other
As long as it’s represented by different float values, that is.
It shouldn’t have been a float to start with in my opinion.
And I’d say
_. -: _.
_.
in keeping with what I had been writing before.
Am 08.02.22 um 20:52 schrieb Raul Miller:
On Tue, Feb 8, 2022 at 2:43 PM Hauke Rehr wrote
don’t see a supporting rationale
for this behavior (but again, I don’t have experience
using it – maybe I just don’t know enough about it?)
Am 08.02.22 um 20:55 schrieb Hauke Rehr:
As long as it’s represented by different float values, that is.
It shouldn’t have been a float to start with in my
Raul Miller:
On Tue, Feb 8, 2022 at 2:55 PM Hauke Rehr wrote:
_. -: _.
Here, we're relying on floating point hardware.
We don't use lookup tables to determine equality -- we use a machine
instruction which tests values for equality.
--
--
mail written using NEO
neo
Just for the sake of pointing out what my idea was,
it would work like this (answers embedded):
Am 09.02.22 um 00:02 schrieb Jan-Pieter Jacobs:
The biggest problem I see is finding a good use for such null value.
In many cases where a null value would make sense, J/APL found a different
way of
answeres below
Am 08.02.22 um 21:12 schrieb Henry Rich:
Generalities may spur the imagination, but to use _. we would need a
detailed spec for what every verb would do given indeterminate input.
Where would it appear in the collating sequence?
Collating:
maybe I’m naive but I thought it
Intel® Core™2 CPU T9550
I understand that may be the problem.
All zeros is quite sane (much better than all ones).
Am 08.02.22 um 21:15 schrieb Henry Rich:
I get 0 0 0 0 0 as I expected. What's your CPU?
Henry Rich
On 2/8/2022 3:07 PM, Hauke Rehr wrote:
As I said, I haven’t been using
I’d love having lua (and metalua), too.
But I also doubt it fits well with J.
Its tables make for a neat data description language, too, imo.
I like them better than the JSON/yaml/whathaveyou style
(whereas for most purposes other than config files, I prefer
either a db or csv files).
Am
namespaces are a general need in programming
objects orientation has its niches where it actually makes sense
even Guido van Rossum’s extremist object oriented language says
$>python
Python 3.9.4 (default, Apr 16 2021, 23:37:41)
[GCC 7.5.0] on linux
Type "help", "copyright", "credits" or
The indices were off by one in my solution wrt your python indices.
You may use these instead:
mat =: (+1=#\)"1 mat
and
mat =: (+3=#\) mat
Am 06.02.22 um 20:35 schrieb Hauke Rehr:
One way to achieve that would be
mat =: (+0=i.@#)"1 mat
and
mat =: (+2=i.@#) mat
Am 06.02.22 um 20:
Where to start?
Maybe work your way up to building that set of indices and
then just use them for amending, maybe something like this
mat =: _ (<"1 ,.~ i. # mat) } mat
for the \ diagonal and
mat =: __ ({ ;~ (#~ 2|>:) i.# mat) } mat
for the even indices
(even wrt python indices? replace >: by ]
One way to achieve that would be
mat =: (+0=i.@#)"1 mat
and
mat =: (+2=i.@#) mat
Am 06.02.22 um 20:24 schrieb Andrew P:
- Add 1 to each value in the first column
x[:,1] += 1
- Add 1 to each value in the third row
x[3,:] += 1
--
--
mail written using NEO
neo-layout.org
2. you could try
echo 'this'
and
smoutput 'that'
3. look up the definition of dyadic # (and $)
(or just try it)
4. I don’t know exactly which behavior you want
could you give some example in- and output
of the more general function you’re after?
Am 27.01.22 um 14:23 schrieb Pawel Jakubas:
Dear
One way to get the simple repeat function is this:
] d =: >: i.3
1 2 3
f =: [ ;@# {.@]
1 f d;d
1 2 3
2 f d;d
1 2 3 1 2 3
3 f d;d
1 2 3 1 2 3 1 2 3
If you don’t call f with d;d but simply with d,
then you can just define
f =: ;@#
which I consider an answer to your question 3.
Is something like this what you want?
repeat =: 4 : 0
rshowx =. 'This is x inside repeat:',LF, ": x
rshowy =. 'This is y inside repeat:',LF, ": y
f =. [;@#{.@]
d =. y
rshowf =. 'These are x and y for f; and its result:',LF,,LF,.":x
(;;f) d;d
rshowx,LF,rshowy,LF,rshowf
)
3
Thanks Julian, this is Good Advice™
My cloumsy answer was meant to say
“that’s not the J way”
without actually spelling that out.
But maybe it didn’t come across.
Sometimes the best answer to “how?” is just “don’t!”
Am 27.01.22 um 15:44 schrieb Julian Fondren:
J doesn't have lexical scope so
I didn’t do it much differently, conceptually.
But instead of binary search (bisection)
I split it up in ten buckets (dekasect).
Apologies for the bad naming:
another greek-latin hybrid, oh my …
I split that way because I wanted to evaluate
ones more often on small inputs than large.
Didn’t
This is a working solution. Not pretty, but it gets the job done.
onesList =: (0<{.)`($:@}. + ((10 ^ <:@#) * 1<{.) + ((1 + 10 #. }.) *
1={.) + <:@# * {. * 10 ^ 2 -~ #)@.(1<#)
ones =: 10 onesList@:(#.^:_1)"0 ]
NB. tests
ones 30 521
13 213
That was easy. Now for the “inverse:”
I don’t quite get it; or rather, I disagree with most of it:
Am 22.01.22 um 14:07 schrieb Raul Miller:
This difference could be prevented by preventing names from being
updated. And, languages have been implemented with that characteristic
(for example, Haskell). (And, of course, quite a lot
mainly @HR
what do you think about
curlyrtcoco =: {{ m x} y }}"0 _
(2 0;1 1) 1 curlyrtcoco i. 3 3
0 1 2
3 4 5
1 7 8
0 1 2
3 1 5
6 7 8
?
Would this be worth adding as a primitive?
I had some use for it, but not all too frequently.
} can be viewed as a three noun argument function,
but you
} would be a more general solution.
Henry Rich
On 1/19/2022 9:53 PM, Hauke Rehr wrote:
mainly @HR
what do you think about
curlyrtcoco =: {{ m x} y }}"0 _
(2 0;1 1) 1 curlyrtcoco i. 3 3
0 1 2
3 4 5
1 7 8
0 1 2
3 1 5
6 7 8
?
Would this be worth adding as a primitive?
I had som
ones1 reads better than the original.
factoring out {. and }. really helps.
even though it means pairs of factors
are “distributed” and one doesn’t
immediately see which one is corresponding.
if it hadn’t been for having it fit on one line,
I had kept the recursive call separate, though.
Am
golfing a bit more, just for fun
(but it may improve performance a tiny bit, too)
ones1=: (1(,,<,=){.) +/@:* (0:`$:@.(0<#),X)@}.
where XX…XX is either of
(*10^<:)@#,(10^#),1+10#.]
(,&1*10^(,~<:))@#,1+10#.]
(,&1*10^_1
But you can. Malbolge no doubt is a language.
Am 22.01.22 um 15:45 schrieb Hauke Rehr:
Wrong. Informed by an implementation means: this is a cooperative
thing so there’s no point in mixing random syntax and semantics.
--
--
mail written using NEO
neo-layout.org
In that case: I stand corrected.
Am 24.01.22 um 06:06 schrieb Elijah Stone:
When I said 'harder to implement performantly', I was thinking of
virtual nouns for verbs _without_ IRS/ITS. But it occurs to me that it
is not actually significantly more work, though you do need a new array
Asked for thoughts, so here they go:
How many operations can be optimized
to not actually bring it to the front?
(special combinations wrt the t. family)
Not that many, I think. So it’s better
to actually
1. prepare your data layout and =: the result
2. apply an easily understood algorithm to
fun game – third try was correct
but I guess the fun’ll wear off soon
which is good because I don’t need another time sink ;)
but there’s an interesting question:
how exactly to prune depending on the feedback?
at first, we don’t want to repeat any letters
we already used in previous guesses
As Henry pointed out,
do take a look at [1] for how to approach this in J
If you want to understand my solution,
just try it step by step.
I repeat it here, with a bit of litter removed:
f =: {{ (,~ {) {. I. (>: x) e."1 +/\ =/~ y }}
NB. where does each number occur?
=/~ d
1 0 0 0 0 0 0 0
Maybe this is what you want (but it’s not elegant)
f =: {{ (,~ {) {. I. ( (>: x)"1) +/\ =/~ y }}
1 f d
2 5
2 f d
3 11
just a quick and dirty working solution
(if 2 5 is the correct answer to 1 f d)
Am 26.01.22 um 19:37 schrieb Pawel Jakubas:
Dear J enthusiasts,
I wonder how to
I don’t understand why it’s index 2 (first occurrence of 2)
but 11 (third occurrence of 3)
Am 26.01.22 um 19:37 schrieb Pawel Jakubas:
Dear J enthusiasts,
I wonder how to elegantly implement dyad f that takes as x=1,2,3... and y
vector and returns nth duplicate value with index of that
rows says: the first entry in col x is the n-th entry in values
row 0 0th (if any)
row 1 2nd (if any)
(so in row 0 we had 2 values)
row 2 3rd (if any)
(so in row 1 we had 1 value)
row 3 5th (if any)
(so in row 2 we had 2 values)
row 4 5th (if any)
(so in row 3 we had 0 values)
etc.
Am 26.01.22
should have read
the first entry in row x …
where x is the index of the number in list rows,
and n is that number
Am 27.01.22 um 01:25 schrieb Hauke Rehr:
rows says: the first entry in col x is the n-th entry in values
--
--
mail written using NEO
neo-layout.org
I’m on a Debian based system and I can’t remember
having installed J using a package manager, ever.
[though maybe that’s only my bad memory]
If there is a package, however, I’d prefer that.
Am 14.01.22 um 15:24 schrieb Raoul Schorer:
Hi,
I noticed that j903 doesn't come as a Debian package
Yes, but this
No, y is ignored, except when it produces an error.
reads like y may produce an error /inside/ rather than before.
As if y would not be evaluated (and maybe throw) outside.
I’d expect that actually y isn’t ignored in that case
but the call doesn’t occur at all because the
talking about conjunctions
@. is much more like conventional if
and works for multiple else-cases (like a switch)
eg
*:`+:`-@.(3|>./)("1) 3 2 $ i. 6
0 2
4 9
_4 _5
Am 13.04.22 um 15:05 schrieb Jan-Pieter Jacobs:
I think your looking for power (^:), a conjunction though.
+: ^:
But Skip seems to want feedback on the question (see subject).
The question is fine, only the answer is flawed.
Flawed in a way to be expected when directing such a question
at Bard.
Am 30.08.23 um 03:13 schrieb Devon McCormick:
Try asking Bard about APL and matrix computations, then point out
why not have a command line argument providing a default number?
$USER may know better what they consider a good default
Am 11.04.22 um 00:31 schrieb Raul Miller:
It would be nice, I imagine, if we eventually had a mechanism like
not too efficient but something to start with
one =: 3 3 $ 'abcdefghi'
other =: 5 6 $ 'abceeedefabcabcdefghidef'
find =: #~ one *./"1@:e.~ , $~ 2 3 ,~ */@:$ % 6:
find other
defabc
abcdef
ghidef
Am 08.04.22 um 16:21 schrieb Richard Donovan:
Hi J programmers
I need help to discover
yes I’m stupid :)
Am 08.04.22 um 16:32 schrieb Henry Rich:
2 3&$"1
--
--
mail written using NEO
neo-layout.org
--
For information about J forums see http://www.jsoftware.com/forums.htm
but with
(9!:19) 1e_11
I get
({.e.}.)"1 w
0 0 0 1 1
Am 17.05.22 um 16:40 schrieb Hauke Rehr:
oops, ignore that one – I see my mistake now
Am 17.05.22 um 16:37 schrieb Hauke Rehr:
I didn’t revisit pe80 (its been quite some time I last looked at pe) but
I get
<./ ({.>./@:|@:-}.)&q
I didn’t revisit pe80 (its been quite some time I last looked at pe) but
I get
<./ ({.>./@:|@:-}.)"1 w
0.539284
so you would need a tolerance greater than that
if you want to get at least one 1
with a sane value for tolerance, a 1 is not to be expected
so what do you want to achieve, what do
oops, ignore that one – I see my mistake now
Am 17.05.22 um 16:37 schrieb Hauke Rehr:
I didn’t revisit pe80 (its been quite some time I last looked at pe) but
I get
<./ ({.>./@:|@:-}.)"1 w
0.539284
so you would need a tolerance greater than that
if you want to get at least one 1
) 1e_11
I get
({.e.}.)"1 w
0 0 0 1 1
Am 17.05.22 um 16:40 schrieb Hauke Rehr:
oops, ignore that one – I see my mistake now
Am 17.05.22 um 16:37 schrieb Hauke Rehr:
I didn’t revisit pe80 (its been quite some time I last looked at pe) but
I get
<./ ({.>./@:|@:-}.)"1 w
0.539284
s
I usually have some similar prelude to some of my scripts.
And I dislike it. I always thought there are better ways
I am unaware of and have been too lazy to look up/ask here.
Am 30.05.22 um 18:37 schrieb Raul Miller:
Well... you could use the approach used here (lines 11 through 17):
the
parts visualizing @ etc to the respective pages, maybe linking back
to an overview assembling all of these illustrations.
Then I’ll be looking forward to your ideas for further improvement.
Eg, I remember Ian not being satisfied with my presentation.
Hauke
Am 28.06.22 um 02:32 schrieb Hauke
(><]) value
can be expanded to
(> value) < (] value)
which yields 0 (“false”) for any unboxed real value
(> (<])) value
can be expanded to
(value > (<])) value
and in turn to
value > (<]) value
and finally to
value > value < ] value
but value < ] value evaluates to the same as (><]) value above
But ←→ is used in maths, too.
In the realm of logic, let φ and ψ be formulæ.
If φ ←→ ψ is a tautology (in semantic parlance “is always true”)
then φ and ψ may always be substituted for each other
regardless of “circumstances” or “surrounding constraints.”
So using that symbol (I don’t know to
If you put something like
export PATH=/home/username/jcurrent/bin/:$PATH
into your .bashrc¹
and keep that symlink jcurrent, well, current,
(which is a good idea in the first place)
then you can just say
#!/usr/bin/env jconsole
instead.
¹ ~ in place of /home/username won’t do in this case!
I’d love to see more hackers here,
and there already are quite some.
the press/the media abused that term
hackers are usually good people
it’s malicious people one has to defend against
That being said, I agree as well.
Am 06.06.22 um 22:07 schrieb Henry Rich:
I agree. If we attract users,
Tom did, in the very post starting this thread:
(1+i. 4 3) +/ . * 1+i. 3 4
38 44 50 56
83 98 113 128
128 152 176 200
173 206 239 272
Am 19.07.22 um 20:10 schrieb Devon McCormick:
I'm puzzled why no one has used the basic matrix multiplication expression
in J.
matmul3
([: +/ *)"1
I agree.
(late to the party, I’ve been away for about a week)
My first approach would have been to let J provide not only
the boxed display (a string) with | replaced by space
but also the tree structure and shapes of representations in boxes.
`$@":`
Then I’d first print the contents with lines
+1 for 'code search' even if it’s a misnomer, strictly speaking
at some point in their J journey, they’ll need/want to try it
and they’ll learn soon enough what it does and doesn’t
that’s my 2¢
Am 29.07.22 um 16:38 schrieb 'robert therriault' via Programming:
Not butting in at all Viktor,
It
An adverb only evaluates once it is given a noun/verb left argument.
(adverb adverb) is a form of speech that doesn’t get evaluated,
just like verbal trains (verb verb verb)
Put another way: what shall 'u' be in advsuc when you say
'advsuc applyto1'? 'u' always comes in from the left
but there
which seems to have been done before
or came up independently
in D, you have slightly more fine-grained options
scope(exit){}
scope(success){}
scope(failure){}
a spontaneous reaction:
In J, I’d like to use generators like
fileworker =. workfile thatone
'exitstatus result' =.
rewrite it
from scratch. :-) )
Ed
*Crutch n. A thing used for support or reassurance. May hold you back in the
long term.
Ed
Sent from my iPad
On Apr 27, 2022, at 7:01 PM, Hauke Rehr wrote:
What keeps you from writing your comments vertically‽
Am 27.04.22 um 14:50 schrieb Ed Gottsman
Smarter than emacs? Vim. :)
editor wars aside, I’m having a hard time
imagining any editor smarter than these two
They may call themselves “smarter” but mostly
that translates to “more cumbersome”.
folds are natural in any coding editor,
and I guess Tim Pope would come up with a solution
Ed,
no apology required :)
For the first approach:
If you want that, you can just start using it.
LEO came to my mind but you could as well
support it in another editor.
Write a function stripping your code off those
comments and feeding the remaining code to J.
Not much of a hassle except for
What keeps you from writing your comments vertically‽
Am 27.04.22 um 14:50 schrieb Ed Gottsman:
If J were written vertically, it might be easier to comment.
--
--
mail written using NEO
neo-layout.org
--
TL/DR version: no, I haven’t
I remember repeated squaring¹ of an integer matrix
where I had wanted to use extended for not losing
the exact precision. Don’t know for sure the results
mostly exceeded 2^63 but I think so.
Those matrices have not been all that “large” however
so maybe this is not
This should obviously be (equivalent but easier to read)
minmax =: r % 4%:2
Am 15.08.22 um 15:34 schrieb Hauke Rehr:
minmax =: 4%: -: r ^ 4 NB. minimum maximal coordinate
--
--
mail written using NEO
neo-layout.org
But in case those “means of nearest points” are actually needed,
the 2D case is easily generated:
require'plot'
r =: 4 NB. some radius
minmax =: 4%: -: r ^ 4 NB. minimum maximal coordinate
X =: minmax+(r-minmax)*1000%~i.1001
Y =: X -~&.:*: X %~ *:r
NB. this is quick and dirty,
works perfectly well here
JVERSION
Engine: j903/j64/linux
Beta-r: commercial/2021-09-26T14:44:24
Library: 9.03.08
Platform: Linux 64
Installer: J903 install
InstallPath: /home/hauke/j903
Contact: www.jsoftware.com
Am 31.07.22 um 13:24 schrieb Razetime:
On this rosetta code page for the
+1 for a landscape version ☺
Am 04.08.22 um 18:18 schrieb Don Guinn:
How about a landscape layout? To see it on a computer screen requires
scrolling unless the text is very small. And the printed version should be
just as easy to use.
1. I found c. easiest to use wrt footnotes
But is this the main question?
I find the other ones easier to use
in general: the footnotes are not
the most frequently looked up parts!
2. c.
3. c.
4. rather c. but I dislike the bold face parts
I could argue about color choice but it’s quite
good
contrary to the answer I gave, I agree:
the names are actually used already so they had better
put on the refcard
And I also agree that some verbs don’t need any explanation.
I thought about putting them in groups more like it’s done
in NuVoc – otoh, if I want to see those relationships, I can
Again, I’d like to see a version of the document,
maybe 1.0.0-draft220809 now.
The headers should align vertically, e.g.
\phantom yScalar monadic verbs\phantom y
(same for Selection, Adverbs etc)
and the footnote mark after Scalar should
in both cases be followed by a ›\ ‹
please use colors
There’s a table for Complex Numbers, and | is \|•\| in that context.
As a Scalar Monadic Verb on non-complex arguments, however,
| is usually learned as |•|, or in J
| -: (* *)
saying | means \|•\| might lead people to believe
5 = | 3 4
even though the heading says /Scalar/ Monadic Verbs
Second
That’s been pretty much negative (but hopefully constructive) critique.
But I also want to thank Viktor et al for the work they put into it.
It will be a good and valuable resource once you decide it’s not a draft
version any more.
Am 09.08.22 um 22:13 schrieb Henry Rich:
Viktor Grigorov has a
+1 for using subscripts, also in explanations of hooks and forks,
and in the Modifier Trains table
regarding footnotes b and c in the Scalar Dyadic Verbs table,
I read c as explaining the /inflection/, and b the attempt at
translating to known semantics in the middle column
So I don’t think it’s
Very interesting indeed, all of this thread.
I think annotation both of nouns and verbs would be useful.
Do we also need a way of de-annotation?
Probably not but I haven’t thought it through.
I don’t like the literal approach. I’d rather have J know about
constants ASSOC. and SORTED. and the
not exactly what I had in mind
so we don’t want no bloat, right
how about simply assigning numbers?
magic numbers are about as readable as magic letters
we already define sugar in english.ijs so some name
could be added for associativity which might be 2⁰=2¹-1
(all four currently known
Nice troll ;)
Generally speaking, in life contexts, “dark” is orthogonal to “useful”.
Which might be related to why Dark Matter is looked up on,
by some in the Physics community — it seems that it Matters too much.
But if the problem is that brightness is insufficiently constrained,
why should
Giving an answer in spite of not having been addressed
because I just criticized the reasoning that I thought
was based on that word’s meaning (dispute resolved privately):
Two vectors are orthogonal if the projection of one onto the other is 0.
In colloquial terms, it means “there is no part of
rank error confirmed
JVERSION
Engine: j903/j64/linux
Beta-w: commercial/2021-12-05T18:24:21
Library: 9.03.08
Platform: Linux 64
Installer: J903 install
Am 25.05.23 um 20:12 schrieb Martin Kreuzer:
Dear all -
Thought I had succeeded in deducing step by step the tacit verb for
factorial,
just a short follow-up
As I know its maintainer in person,
I had to try GeSHi as well.
Not overwhelming, but working fine ootb.
Am 30.07.23 um 19:42 schrieb Hauke Rehr:
my 2 cents
I’m quite satisfied with the highlight I get when
writing J in Vim; and with the highlight I get when
typesetting
my 2 cents
I’m quite satisfied with the highlight I get when
writing J in Vim; and with the highlight I get when
typesetting J in TeX with minted. Neither do custom
handling of Special Combinations, though.
IIRC, Michal rolled his own for his jprez stuff
(editor part).
@Michal: Did you consider
Sorry, maybe was there a wrong replyto address,
or maybe I just mistyped.
Weitergeleitete Nachricht
Betreff: Re: [Jprogramming] Differentiation
Datum: Sat, 22 Jul 2023 18:05:02 +0200
Von: Hauke Rehr
An: programming@forums.jsoftware.com
In other words,
a =: 5
You are right, C is an ususable language.
I think noone uses it. You cannot search.
This thread is silly.
J is one of single letter names.
It’s fine.
Am 09.05.23 um 22:56 schrieb Richard Donovan:
One of the most frustrating things about J is its name! For example, on Stack
Exchange code
invented this
remarkable language didn’t foresee this problem in the first place!
On 9 May 2023, at 22:47, Hauke Rehr wrote:
You are right, C is an ususable language.
I think noone uses it. You cannot search.
This thread is silly.
J is one of single letter names.
It’s fine.
Am 09.05.23 um
In case you are still using a badly designed keyboard.
Am 10.05.23 um 16:34 schrieb Michael Dykman:
I second Harvey's suggestion: 'jkl'
It is very easy to type
On Wed, May 10, 2023, 04:14 HH PackRat wrote:
I may be completely mistaken (and I didn't have time to double check
the history),
No, I don’t think there’s as much junk regarding J.
Am 10.05.23 um 18:51 schrieb Raul Miller:
Has anyone written that content for the J community?
Thanks,
--
--
mail written using NEO
neo-layout.org
--
Dear God, why did you take the name of “Richard Donovan”
to tell me about your omniscient wisdom regarding other
people’s ignorance?
Am 10.05.23 um 19:02 schrieb Richard Donovan:
They haven’t realised the benefits yet!
From: Programming on behalf of Hauke Rehr
Thanks for the information. Much appreciated.
I didn’t get the chance to converse all that much with him.
Still, I miss his humor.
On 29/12/2023 18:07, Eric Iverson wrote:
Don Knuth's current work in progress is: "The Art of Computer Programming
Volume 4 Fascicle 7" with publication expected in
301 - 400 of 400 matches
Mail list logo
