Delusional statements in article comments in NI.com

[altenbach]

In terms of recreating old technology, I got a kick out of a full Cray-1 implementation using a cheap FPGA board. 😄

Back to the original topic:

• The comment:

The quoted post comment actually has very little informative content. 90% is just vitriol, and 8% incorrect statements and 2% half-truths.

I agree with Saverio that censorship needs a very high bar in general, but this is a company web site and certain things just don't belong here. Other sites have problems with spam posts or irrelevant posts with links (e.g. just to bolster search ratings), for example. There is no questions that these should be deleted as soon as they appear. I think any anonymous post should have a lower bar than posts with an author name. Replying with another comment is just feeding the trolls.

Another issue is with personal attacks. Calling NI tech writers "self appointed experts that are not good with computers and should get another hobby" is a bit much.

Narrowly focusing on bits or other countable features for any kind of quantification is silly.  A spider with 8 legs is not twice as fast as a cheetah with four legs. An 18 wheeler is not 9x faster than a motorcycle unless you measure the sustainable rate of transporting palettes of bricks from point A to point B.  😄

• The actual article:

Yes, there are simplifications and incorrect statements, but there is no way to condense this into a readable 1 page article without taking some shortcuts.

A simple example that I already mentioned:

QUOTE: " The bit size of a processor refers to the size of the address space it can reference."

Disproval: All we need is finding a single exception to prove the statement wrong. Many 8 bit processors have a 16 bit address bus and can thus address much more than 256 bytes. Clearly the quote is thus incorrect.

We have N-bit registers, N-bit data buses, N-bit address buses, N-bit SIMD operations, N-bit coprocessors (e.g. for encryption), N-bit OS, N-bit applications, physical address extension, motherboard limitations, etc. etc. and nailing it down in a few sentences is not trivial. Somebody could write a tome and it still might be flawed because of missing consesus among experts on how to categorize things.

An introductory article such as this on ni.com should narrowly focus on the differences, implications, and caveats of 32 bit vs 64 bit LabVIEW to give the undecided reader an informed choice and maybe some simplified background. I think the current article mostly succeeds here.

Fortunately, in a few years all 32 bit OS for desktops will be extinct, it will only survive for a few more years on phones and other simple gadgets. The cost of RAM is such that even budget PCs and laptops now have more than 4 GB of RAM, absolutely requireing a 64 bit OS.

• So what is the solution?

Documents on the NI community pages offer options that articles can be edited by anyone (wikipedia style). Would that be a better options?

[Ben]

I have no idea what happened above.

I did try to say that...

even a pdp-1170 (circa 1978) had 22 bits of addressing and 16 bits of data.

Ben

[altenbach]

---

@Ben wrote:

even a pdp-1170 (circa 1978) had 22 bits of addressing and 16 bits of data.

---

Back in the days, I did a lot of Fortran programming on a similar unit using a tektronix vector terminal. 😄

(Editing was all key command driven, entries terminated by "escape-escape" if I remember right.)

[aeastet]

I love it when you guys start showing off all of the stuff you used to progtram with. It makes me feel so much younger.

[LV_Pro]

Christian, I will try to give you proper attribution when I try to work this into conversation "A spider with 8 legs is not twice as fast as a cheetah with four legs. An 18 wheeler is not 9x faster than a motorcycle unless you measure the sustainable rate of transporting palettes of bricks from point A to point B."! It made me laugh out loud. Thank you for a good start to my day!

There always is a tendency for critical reviews of technical statements in the press, for there imprecision, if not total innaccuracy. There is also a trend of psuedo-experts, or self appointed experts, whose own knowledge of  "facts" may be flawed, to comment on these topics. It has been very difficult for me to remain silent when cruising the aisles at (name your electronics retailer) when I hear some sales person spouting off to their prosepctive customer about some technical aspect of (pick your own consumer electronic product), for which they are obviously in error. I see this situation, and others like it, as just a related phenomena.

As to censorship, I agree that the bar should be high, otherwise half of what I say would be missing! 😉

[cajacobs]

Laura,

It would be nice if a direct reply to a particular thread post were indented under it, and you could easily see which posts are NI endorsed and which are from the user community as a whole.

[Norbert_B]

This is really a tough discussion topic: is censorship for a webpage like ni.com OK or not?

Gagging people for their thoughts is a no-go for a community considering itself "open-minded" (which i would count ni-community). This arises the issue that made comments can be read and "evaluated" for everyone.

In the quoted case, i would think that it is quite obvious, that the poster is in fact a troll. Otherwise, he would have stated his contact or at least his name for further discussion.

And this leads to my proposal:

Why do we not remove the "comment" area from white papers at ni.com, but insert a link to a discussion thread on this forum? So introduce a new folder "Article discussion" and each new article posted creates a new discussion thread automatically. Compare it like the "Monthly bug report" threads. This enables the community to discuss different views on the topics.

thanks,

Norbert

PS: It is rather "dangerous" for most of us to stand up and write such an article. I have this feeling since most processes and technologies are so complex, that nobody knows everything about this. So you will always (in addition to the reasons altenbach already posted like e.g. "simplicism" for improved readablity for non-experts) run into the situation that you make mistakes. Nevertheless, i am glad that people work hard for creation of such documents since even if there are errors in there, it is better than having no document at all!

[RayFarmer]

One problem I could see with linking it to a forum discussion, is that quite often discussions do branch off the topic and sometimes never really get back on the topic.

When comments are made to white papers or web pages, are they not checked for suitability for publication before appearing on the website?

[PhillipBrooks]

You must have an NI profile to post a comment on a KB article.

Each time you post a comment on the KB articles, the following is displayed 

---

The Fine Print:

• All submitted reviews become the licensed property of National Instruments Corporation as set forth in our Legal Notices.
• Your reviews will be posted within five to seven business days.
• Submissions that do not follow our review guidelines will not be posted.

---

 

The review guidelines state:

---

Any comments in violation of these guidelines, or the site's terms of use, may not be posted.

Although you must provide your NI user profile, you can opt not to display the information with your comments. All comments are reviewed by an editor and are subject to editing before being posted. The comments are posted within five to seven business days.

Finally, we respectfully request that you refrain from including the following in your comments:
 

• Profanity, obscenities, or spiteful remarks.
• Single-word comments. We want to know why you liked or disliked the information in the document.
• Advertisements or offers to sell or buy any goods or services for any business purpose.
• Phone numbers, mail addresses, or URLs.

---

NI could remove the post, edit the post, or leave it stand. I would personally label the comment as spiteful, remove it and notify the originator that the post was rejected, optionally with a reason why.

It appears to me that NI is not reviewing the posts and is allowing the posts to be added immediately.