Indeed. The suits get nervous and catch featuritus. In my opinion Youtube should improve its existing commenting feature. It's hard sift and re-find comments if there's a lot per vid, for example. And the reply notifier is buggy.
Forget about being a blog/photo/texting gizmo. You lost that battle already.
I'd LOVE to be trollin' around slashdot etc. instead of actually driving during my commute.
And my lazy son won't get a driver's license, making me drive. I'm thinking of telling him to Uber, but the idea of riding with a creepy stranger kind of bothers me. I'd rather it be creepy robot.
There are people too young, too old, or too ill to drive, and many that just don't want to. I'd say that's at least 1/4 the population (excluding younger than say 12). Big potential market.
The system doesn't have to be perfect, only be equal to or better than typical human taxi drivers. Human drivers make mistakes and/or have bad maps also.
you don't know what the fuck OO is in the first place
Nobody agrees on the definition or formal characteristics.
And nobody agrees on who gets to decide the definition. There is no standards body who defines what OO is. YOU are not it.
Every tool has a proper place and use, and making relative value judgements between them is something fuckwits do when they only know how to use one thing.
I've mentioned multiple times I do choose to use it to some degree. I've found practical utility for it under certain circumstances. So what is this "one thing"?
how you've rooted out this terrible OO hoax perpetrated on the industry
Well, okay, "hoax" is possibly a little harsh, but I do believe that OO-domain-modeling was thrust on the industry by hucksters wanting to sell books and consulting services; and millions, if not billions were spent chasing this false dream.
The primary focus of these hucksters was in domain modeling. Therefore, other than a small blurb, I didn't used to make much of a distinction between "OO" and "OO domain modelling".
A bigger distinction is made now in the industry because practical experience and failures taught many that domain modeling is NOT where OO shines. I know you disagree that domain modeling was the main stated selling point of OO, but we'll just have to let that disagreement stand (unless somebody can present reliable surveys of beliefs).
Why does it matter to you anyhow? It's water under the bridge. I'm not understanding your complaint. It appears you want to make this all about me instead of OO. That's why you appear to be a troll. Non-trolls talk about the subject at hand. Say something about OO and than back your claim with evidence: that's not asking too much.
You seem to agree there that scientifically comparing the utility of software engineering paradigms/techniques is either a gray art or requires resources not yet committed by anybody. Good, that part is mostly settled then.
Anyone who has ever sent an email knows there are at least 2 copies. One on the sender's account. One on the recipient's. If anyone else is CC'd, then they have a copy too. Did anyone believe when she 'wiped her server' (even without a cloth), that they all disappeared forever?
I've read that the "regular" office server croaked and the backups were fucked. Hillary and Colin were correct about one thing: gov't IT sucks.
I tried to find links to post about it, but the name-space is too flooded with articles about H's server now. I'll keep looking...
My interpretation was that she was saying a category classification of "terror" versus "non-terror" was a premature and/or irrelevant question JUST right after the attacks had taken place. The terror/protest dichotomy was something the news cycle created, and is possibly useless (especially being it was likely a combination of both: a smaller plan expanded by the existence of protesters).
But, pundits frame it as a summary dismissal of the question.
It appears he did make a general recommendation related to an outside account, but perhaps Clinton's aides over-spun it, not necessarily Hillary herself.
She has been legally and politically pressured to describe in detail why she didn't use the gov't system, and recommendations from Powell would be a legitimate part of that.
First you present your SLM evidence that it's usually better.
A better source of my thoughts is probably the c2.com wiki. Discussions from many about many software engineering topics, including OO, are there. It's in read-only now, though, due to vandals.
Comcast is delusional if it thinks speed is the major bottleneck between subscribers and happiness. It is but one of many issues...
Most services typically face periods of sluggishness. Based on my experience with slimy telecom marketers, they'll suggest that those periods are due to not having high enough "speed". Naive customers may just fall for it.
In reality, those periods of sluggishness will likely happen regardless of "billed" speed. I've purchased higher speed mainly because it was bundled* with something else, but I still get periods of sluggishness, especially on weekend evenings. They all oversell capacity, probably with DSL also. (It's not Comcast, but I've read similar complaints from users of ALL the big telecom's. Oligopolies usually suck, period.)
Thus, speed is probably being used as a sales gimmick.
* I hate forced bundling, but that's a different evil-telecom subject.
I believe we need something that's a combination of file system and database. I've kicked around ideas for such. Generally we need hierarchies for "regular" people to relate to stuff, but sometimes we need a database-like layer or activity on files.
I've kicked around the idea of using Extended File Attributes, along with "dynamic relational". Basically you'd define the file path as a virtual table, then query that table using a dynamic version of SQL. Pseudo-code:
define table foo /networkX/zib/flahg/*.pdf /networkY/gerf/*.* end define begin query fliggo
SELECT * FROM foo
WHERE filename LIKE "library%"
AND modify_author IN ('Bob','Martha','Suzan') end query DISPLAY filename, modify_date, path
FROM fliggo
In 15 years of working with OOP I have never heard it sold primarily on the basis of domain modeling...
My experience differs. It's quite possible the presenters were naive or "unqualified", but I'm not judging the quality of such claims/claimers in that spot, only their commonality. You tend to over-interpret specific statements of mine.
nobody had mentioned domain modeling until you showed up
There's a lot things they didn't mention. That's hardly a decent metric for industry presentation quantity.
because to be honest everything you have said so far has been nonsense to me.
The feeling is mutual.
Would that be the 'real science and real numbers' which you directed me to acquire earlier without citing any references?
It's purely my observational opinion. We don't have solid science/math/logic either way, so it's your word against mine.
makes me feel absolutely confident in ending a candidate interview immediately
I'm not here to please you and fortunately don't have to.
Until somebody has solid science/logic/metrics* (SLM) to show that OOP is better for most or all parts of applications, it's merely opinions and anecdotal evidence. If you want to insist it's better, then find a way to present your case well rather than imply that skeptics are of low skill/intelligence. Makes you sound like a damned troll.
Had you stuck around long enough to learn design patterns,
The design pattern case was poorly presented. They should have presented a variety of implementation choices and come up with clear SLM to show that their implementation choice is the "best" choice. Many seemed like arbitrary implementation decisions to me. Often such choices will depend on the context or the domain and/or usage of the tool.
without mentioning a single actual OOP principle.
A principle is not necessarily a metric. Satisfying Principle X doesn't necessarily make a tool overall better. Generally one will be weighing multiple principles against each other because it's usually not possible to optimize them all at once.
No escaping for the need for SLM.
Humans do not think in compartmentalized regions of internally consistent logic assembled into 'correct by construction' higher level constructs. They are irrational and perfectly happy to proceed with actions and assumptions while lacking proof that they are valid or even while staring contradictory evidence in the face.
Maybe, but if Ferengi's are going to be maintaining the code in the future, it may be a bad mistake to code for Vulcans. Car co's design cars for regular folk, not Indie 500 drivers. Perhaps you are making unrealistic staffing assumptions.
* Beyond narrow/specific metrics, such as "scores high on the Fiddlebop Reference-Cycle metric. Too many people are happy to judge something on narrow or few pet metrics.
by making the utterly ludicrous statement that domain modeling was the main thrust of OO.
It was the main thrust as promoted and taught. I used to blog heavily on OOP and followed that stuff closely. ("Reuse" was a close second, the 3rd being "fitting the way humans think", by my estimates.)
Now whether that "should be" the main thrust is another matter, for I'm just reporting on human interest level in that spot; not tool value.
The "value" issue gets back to the science and evidence debate: proving objectively that X is better than Y rather than "trust me, I'm smart" (beyond one or two narrow factors.) The opinion-to-science ratio is so very high in the OO biz. (Well, software engineering in general. Good SE science is expensive and nobody wants to pay for it.)
The idea is sound: only give people access to the systems and documents they need access to. The problem is that you'll never know beforehand which systems and documents those are. So, you need access to Doc X? "I know it takes 2 weeks to process an access request for this folder, so why don't I just email you the thing.
Mozilla is soliciting comment and criticism on the seven new designs... but this is no Boaty McBoatface situation. Mozilla is clear that it's not crowdsourcing a design...
OSS has tools with names such as Gnu, Gimp, Postresql, Vuze, Ogg Vorbis, and Troff. I don't see Boaty McBoatface being any worse.
Names that sound like alien medical conditions and bodily fluids actually seem to give OSS tools "street cred", due to being names corporations would typically reject. They are nerdily refreshing after dealing with names like "Excel", "Power Point", "Share Point", "Synergy Dealmaker Pro" (okay, I made the last one up, but I bet The Suits floated it.)
Slashdot Mirror
Hindenbutt: Oh the huge-fannity!
Indeed. The suits get nervous and catch featuritus. In my opinion Youtube should improve its existing commenting feature. It's hard sift and re-find comments if there's a lot per vid, for example. And the reply notifier is buggy.
Forget about being a blog/photo/texting gizmo. You lost that battle already.
I'd have to disagree.
I'd LOVE to be trollin' around slashdot etc. instead of actually driving during my commute.
And my lazy son won't get a driver's license, making me drive. I'm thinking of telling him to Uber, but the idea of riding with a creepy stranger kind of bothers me. I'd rather it be creepy robot.
There are people too young, too old, or too ill to drive, and many that just don't want to. I'd say that's at least 1/4 the population (excluding younger than say 12). Big potential market.
The system doesn't have to be perfect, only be equal to or better than typical human taxi drivers. Human drivers make mistakes and/or have bad maps also.
Just tell Republicans that Hillary wrote it and a warehouse full of lawyers and clerks will analyze every character.
...mutations propelled life.
You are full of stuff.
Nobody agrees on the definition or formal characteristics.
And nobody agrees on who gets to decide the definition. There is no standards body who defines what OO is. YOU are not it.
I've mentioned multiple times I do choose to use it to some degree. I've found practical utility for it under certain circumstances. So what is this "one thing"?
Well, okay, "hoax" is possibly a little harsh, but I do believe that OO-domain-modeling was thrust on the industry by hucksters wanting to sell books and consulting services; and millions, if not billions were spent chasing this false dream.
The primary focus of these hucksters was in domain modeling. Therefore, other than a small blurb, I didn't used to make much of a distinction between "OO" and "OO domain modelling".
A bigger distinction is made now in the industry because practical experience and failures taught many that domain modeling is NOT where OO shines. I know you disagree that domain modeling was the main stated selling point of OO, but we'll just have to let that disagreement stand (unless somebody can present reliable surveys of beliefs).
Why does it matter to you anyhow? It's water under the bridge. I'm not understanding your complaint. It appears you want to make this all about me instead of OO. That's why you appear to be a troll. Non-trolls talk about the subject at hand. Say something about OO and than back your claim with evidence: that's not asking too much.
You seem to agree there that scientifically comparing the utility of software engineering paradigms/techniques is either a gray art or requires resources not yet committed by anybody. Good, that part is mostly settled then.
Roughly 50 Clintonian conspiracies, and so far no smoking gun.
Quiz: Occum's Razor would select:
A) Clintons are master criminals
B) GOP and conserv. media are full of it
Note that "A" also contradicts being sloppy with emails.
Another link:
http://www.nytimes.com/2016/08...
Evidence, please. A bimbo saying so on Fox News is not evidence. (Pleasant to watch perhaps, but that's another issue.)
I made up science? Just call me Francis Bacon Jr.
Damned troll!
I've read that the "regular" office server croaked and the backups were fucked. Hillary and Colin were correct about one thing: gov't IT sucks.
I tried to find links to post about it, but the name-space is too flooded with articles about H's server now. I'll keep looking...
Conservative spinners keep spinning the "What difference does it make" comment. Here's more detail and quotes:
http://www.politifact.com/trut...
My interpretation was that she was saying a category classification of "terror" versus "non-terror" was a premature and/or irrelevant question JUST right after the attacks had taken place. The terror/protest dichotomy was something the news cycle created, and is possibly useless (especially being it was likely a combination of both: a smaller plan expanded by the existence of protesters).
But, pundits frame it as a summary dismissal of the question.
I don't know that's an accurate way to describe it. The truth may be more nuanced:
http://www.cbsnews.com/news/co...
It appears he did make a general recommendation related to an outside account, but perhaps Clinton's aides over-spun it, not necessarily Hillary herself.
She has been legally and politically pressured to describe in detail why she didn't use the gov't system, and recommendations from Powell would be a legitimate part of that.
Sarcasm won't fix your lack of SLM.
First you present your SLM evidence that it's usually better.
A better source of my thoughts is probably the c2.com wiki. Discussions from many about many software engineering topics, including OO, are there. It's in read-only now, though, due to vandals.
Most services typically face periods of sluggishness. Based on my experience with slimy telecom marketers, they'll suggest that those periods are due to not having high enough "speed". Naive customers may just fall for it.
In reality, those periods of sluggishness will likely happen regardless of "billed" speed. I've purchased higher speed mainly because it was bundled* with something else, but I still get periods of sluggishness, especially on weekend evenings. They all oversell capacity, probably with DSL also. (It's not Comcast, but I've read similar complaints from users of ALL the big telecom's. Oligopolies usually suck, period.)
Thus, speed is probably being used as a sales gimmick.
* I hate forced bundling, but that's a different evil-telecom subject.
I believe we need something that's a combination of file system and database. I've kicked around ideas for such. Generally we need hierarchies for "regular" people to relate to stuff, but sometimes we need a database-like layer or activity on files.
I've kicked around the idea of using Extended File Attributes, along with "dynamic relational". Basically you'd define the file path as a virtual table, then query that table using a dynamic version of SQL. Pseudo-code:
(Slashdot messes up my indents.)
My experience differs. It's quite possible the presenters were naive or "unqualified", but I'm not judging the quality of such claims/claimers in that spot, only their commonality. You tend to over-interpret specific statements of mine.
There's a lot things they didn't mention. That's hardly a decent metric for industry presentation quantity.
The feeling is mutual.
It's purely my observational opinion. We don't have solid science/math/logic either way, so it's your word against mine.
I'm not here to please you and fortunately don't have to.
Until somebody has solid science/logic/metrics* (SLM) to show that OOP is better for most or all parts of applications, it's merely opinions and anecdotal evidence. If you want to insist it's better, then find a way to present your case well rather than imply that skeptics are of low skill/intelligence. Makes you sound like a damned troll.
The design pattern case was poorly presented. They should have presented a variety of implementation choices and come up with clear SLM to show that their implementation choice is the "best" choice. Many seemed like arbitrary implementation decisions to me. Often such choices will depend on the context or the domain and/or usage of the tool.
A principle is not necessarily a metric. Satisfying Principle X doesn't necessarily make a tool overall better. Generally one will be weighing multiple principles against each other because it's usually not possible to optimize them all at once.
No escaping for the need for SLM.
Maybe, but if Ferengi's are going to be maintaining the code in the future, it may be a bad mistake to code for Vulcans. Car co's design cars for regular folk, not Indie 500 drivers. Perhaps you are making unrealistic staffing assumptions.
* Beyond narrow/specific metrics, such as "scores high on the Fiddlebop Reference-Cycle metric. Too many people are happy to judge something on narrow or few pet metrics.
Better yet, hire somebody desperate but eager from the 3rd world for whom $500 means living well for a year...or even just living for a year.
It was the main thrust as promoted and taught. I used to blog heavily on OOP and followed that stuff closely. ("Reuse" was a close second, the 3rd being "fitting the way humans think", by my estimates.)
Now whether that "should be" the main thrust is another matter, for I'm just reporting on human interest level in that spot; not tool value.
The "value" issue gets back to the science and evidence debate: proving objectively that X is better than Y rather than "trust me, I'm smart" (beyond one or two narrow factors.) The opinion-to-science ratio is so very high in the OO biz. (Well, software engineering in general. Good SE science is expensive and nobody wants to pay for it.)
Hillary, welcome to Slashdot! ;-)
OSS has tools with names such as Gnu, Gimp, Postresql, Vuze, Ogg Vorbis, and Troff. I don't see Boaty McBoatface being any worse.
Names that sound like alien medical conditions and bodily fluids actually seem to give OSS tools "street cred", due to being names corporations would typically reject. They are nerdily refreshing after dealing with names like "Excel", "Power Point", "Share Point", "Synergy Dealmaker Pro" (okay, I made the last one up, but I bet The Suits floated it.)
Wrong. My very first message I made a distinction. Read better.