Jump to content
DavidHertzberg

PSA: Wikipedia article on Retrospect going away in current form

Recommended Posts

What I see here is:
- Someone is attempting to use WikiPedia as a wiki platform for writing Retrospect user-generated info of various kinds

I would suggest backing off from that. As others have noted, Wikipedia isn't a particularly trusted platform anyway, and they dislike misuse/abuse of the platform.

Alternatives include:

- Perhaps Retrospect could implement wiki.retrospect.com

- Or, just make a shared Google Drive folder. Easy peasy.

Share this post


Link to post
Share on other sites

wiki.retrospect.com sounds like a good idea. User contributed articles can be aggregated in a single place instead of scattered across the forums. Combine it with the forums and the same login could be used for both.

The problem I see with Google Drive is for the users who have no prior investment in that ecosystem.

Share this post


Link to post
Share on other sites

MrPete,

Thank you for trying to help.  However the criticisms you raise have been already dealt with extensively in the preceding posts of this thread.

DovidBenAvraham, back in the fall of 2016, was indeed "attempting to use WikiPedia as a wiki platform for writing Retrospect ... info of various kinds", but the info was originally mostly from Retrospect Inc. documents and was intended as a concise introduction to Retrospect's features—historically organized.  By the fall of 2017 other Wikipedia editors were strenuously objecting to the article's length—originally 9.5 screen pages—and its user-generated "inferences and asides" about inadequacies in the User's Guides.  DBA had already eliminated these "inferences and asides" by the time the OP in this thread was written, cutting the article to a 7.5-page version preserved here

This was still too long and too much like a user's guide for other WP editors, so in November 2017  DBA ditched the historical organization and—per this post above—split the article into two articles.  The "Retrospect (software)" WP article is less than 2 screen pages.  The second article is a a new section at the back of the "Backup" article, and is written so as to describe features common to all enterprise client-server backup applications.  That section contains links to descriptions of corresponding features in two other enterprise backup applications, hereinafter referred to as NB and BE, and is 2.2 screen pages long.  As for "misuse/abuse of the platform", DBA has conducted painstaking discussions of the contents of these articles—which you are welcome to read on their WP Talk pages.  It suffices to say that there have been no objections from other WP editors to any Retrospect-related item DBA has written since March 2018.

That's not to say that there hasn't been "misuse/abuse of the platform" in the WP articles about the other two enterprise client-server applications mentioned in the preceding paragraph.  The NB article isn't too long; it's only 2.5 screen pages—and was 2 screen pages in 2015.  The BE article, however, is 6.5 screen pages long; it was 2 pages long in 2015.  The reason I mention 2015 is that it seems to have been the time at which a new spirit took over at Wikipedia, trying to make it a "particularly trusted platform" by demanding more explanatory links and more references in articles in place of what has been called Making Stuff Up from industry-transmitted wisdom.  The two articles DBA wrote were obliged to comply with the new spirit, but IMHO the NB and BE articles seem to have been "grandfathered in"—and are rather incomprehensible to anyone not already familiar with the terminology of the applications they describe.

Exclusive of immediately after a revision to the "Retrospect (software)" article has been made—when it seems as if the entire working population of Walnut Creek CA wants to read it, that article gets an average of 20 views per day.  The BE article gets an average of 100 views per day, as does the NB article.  DBA and I would like to believe that the 20 views per day represent potential customers for Retrospect.

 

Edited by DavidHertzberg
In 4th prgf., changed "Making Things Up" to "Making Stuff Up"

Share this post


Link to post
Share on other sites
On 10/18/2018 at 4:50 PM, Scillonian said:

wiki.retrospect.com sounds like a good idea. User contributed articles can be aggregated in a single place instead of scattered across the forums. Combine it with the forums and the same login could be used for both.

The problem I see with Google Drive is for the users who have no prior investment in that ecosystem.

Scillonian and MrPete,

Let's start with two simple questions about user-contributed articles:

  1. What kind of content will be allowed in these articles, and who is going to be the moderator(s)?
  2. What makes you think Retrospect Inc. will pay for this, or support it on this website?

Before anyone starts answering these questions, he/she should carefully read this section of the "Retrospect (software)" article—preferably including the referenced articles by DeLong 2012 and Engst 2009 and Friedman 2010 (currently refs 5, 11, and 12) , and should read between the lines applying any experience in an organization he/she may have had.  The conclusion he/she will undoubtedly reach is that anyone working on Retrospect for EMC/Roxio/Rovi from 2007 through 2011 suffered through a near-death employment experience (The Register author Ashlee Vance went on to write for the New York Times) ,which seems to have left those people—many of whom now work for Retrospect Inc.—with what I would call a form of PTSD.  (If you think this is too strong a statement, notice that the head of Retrospect Tech Support—an employee since 1994—posted this link to the DeLong 2012 article as the second thread in the newly-established Latest News forum in 2012.)

What that means is that IME Retrospect Inc. has absolutely no tolerance for criticism of Retrospect Inc. or its products on this website.  An example from October 2017 is a post of mine the head of RTS deleted, referred to in the parenthesized last sentence of the first paragraph of this post.  The sarcasm referred to was a clumsy riff on the "flying saucers" reference in this previous post in the same thread, which IMHO should have easily been understood by anyone with a sense of humor as a joke.  It wasn't, and that's my one occurrence of "abusive behavior".  If the head of RTS doesn't in fact have a sense of humor, how do you explain this previous post in this thread—especially given my reply two posts below it?  I think my "abusive behavior" was really an easily-understood too-sharp reference to the "famous Documentation committee".  I've been more careful to avoid criticism in my posts ever since, which is why I don't think any user-contributed articles that are at all critical (other than bug-reporting posts that are now ignored unless turned into Support Requests) will be allowed on the Retrospect Inc. website—or paid for by Retrospect Inc. on another website.

Why do you think Retrospect Inc.'s Support Request system doesn't allow any customer except the one who submitted a Support Request to see it?  I know that the people in Retrospect Sales have such a capability, and so do customers using other software companies' equivalent systems.  IMHO the reason is that, stemming from its employees' collective PTSD, Retrospect Inc. doesn't want to take the chance that potential customers—or even existing customers—might see how many existing bugs there are.  That might, in Retrospect Inc.'s view, also amount to criticism of its products and/or organization.

So what kind of user-contributed content, other than what is already on the Forums or in the Wikipedia articles, does that leave?  We could have systematic discussions of information about features that are planned or not fully operational, which DovidBenAvraham is not allowed to put in the WP articles.  But we already have such discussions on the Forums; I've contributed to a few of them on the Product Suggestions Forums and elsewhere.  In fact the second paragraph of the preceding post in this thread I've linked to two sentences above is an attempt to start such a discussion about the beta Storage Groups feature.  But we don't get any participation now in those discussions from anyone working for Retrospect Inc., per a statement e-mailed to me by JG Heithcock which I first quoted in the second paragraph here.  That wouldn't change with user-contributed articles, so again I don't see any advantage to be gained.

Edited by DavidHertzberg
Added fifth paragraph; in 2nd prgf. added link to The Register story about mass layoffs at EMC Walnut Creek

Share this post


Link to post
Share on other sites

DovidBenAvraham has now manged to do two things:

First, by adding a second paragraph to this sub-section of the WP "Backup" article, he has managed to describe snapshot-based backup applications as a variant of enterprise client-server backup applications.  That has allowed him to put in references to R.V., although the WP rules for that article do not permit him to actually mention the name R. V..  Moreover DBA has actually managed to mention the existence and name of R. V. in this section of the "Retrospect (software)" article, although he was obliged to say "that separate product, with totally different code,  is not discussed in this article."  Incidentally the non-marketing documentation for R.V., consisting of a handful of YouTube videos, is so sparse as to make it impossible for DBA to discuss it on WP.  Somebody in Retrospect Inc. Marketing had better get a third-party author to write a review, but the review can't be on TidBITS because R.V. doesn't backup Macs.

Second, based on the discovery by JamesOakley that "under 15.x for Windows, even Desktop edition gets multiple execution units,"  I was able to verify that the same is true for "activity threads" under Retrospect Mac 15.0—using the test version Retrospect Tech Support had given me for diagnostic logging of -530 bugs.  (I still run production backups using the Retrospect Mac 14.6 "backup server" and the 14.1 Client, because later versions disable my -530 workarounds.)  DBA has now managed to work an announcement of that feature into this section of the "Retrospect (software)" article.  The new feature, of course, is simply a disabling of the Desktop Edition code that previously reset the number of units/threads to 1 every time the Engine was started—thus forcing Desktop Edition to mimic the non-multithreaded Retrospect Windows/Mac 6.  :rolleyes:  Retrospect Inc. is apparently so reluctant to mention this improvement that DBA had to reference it with a roundabout bug fix note in the cumulative Retrospect Windows Release Notes. <_<

Share this post


Link to post
Share on other sites
On 3/20/2018 at 4:54 PM, Scillonian said:

It'll be interesting to see how they will define what are 'server-level Linux distributions'.

Motivated by the discovery that Version 15 has removed the disabling of multi-threading for Desktop Edition, I started this thread on the Ars Technica Linux Kung Fu forum to ask about the technical feasibility of:

  • creating a Retrospect Client for Linux x64 Server that will run on what it determines is a server distribution
  • making the Retrospect Client for Linux x64 unable to run on what it determines is a server distribution
  • having the Retrospect Client for Linux x64 Server identify itself to the "backup server" as requiring an additional expensive license code

You can look at the linked-to Ars Technica thread for the technical outline of methods other posters suggested for how a Retrospect Client could determine whether it is running on a server distribution of Linux.  However one poster replied:

Quote

In the Linux world, there's a wide variety of stuff that might be called a "server", so I'm not sure market segmentation based on server/client makes much sense. Say,

a) A monster box sitting in some data center. - Yes, a server.

b) A small NAS box sitting in somebody's closet - also a server.

c) A container image, one of dozens or even hundreds running on the same piece of hardware. - also a server.

Most modern distros have an os-release file in a standard location and format which describes, among other things, the name and version of the distro.

My tentative conclusion takes into account that Retrospect Inc.'s problem would be customers who fall into that poster's category b). Is that "small NAS box sitting in somebody's closet - also a server" being used merely by a multi-computer family that just wants to store still pictures and videos etc. in a common accessible location? Or does someone in the family also/instead run a small business administered using more than one computer? In the first case, the family certainly would legitimately argue that it shouldn't pay an extra US$500 for a Server backup license. In the second case, the family might also argue that, because of the size of the business, it shouldn't pay an extra US$500 for a Server backup license.

However it also takes into account a response from another poster saying "Debian is frequently used on servers, but I have also used it as a desktop system. So that OS file may not really be of help."  I conclude that Retrospect Inc. may have now realized engineers would at the minimum be required to continually update the Retrospect Client for Linux x64 Server code to keep up with changing Linux distributions.  Moreover IMHO engineers would want to "fail safe" by treating a distro Retrospect Client for Linux x64 Server couldn't recognize as one for which it needn't charge extra, rather than "fail unsafe" by insisting on charging extra to back up that client.  Whether Retrospect Inc. will try to implement charging extra for backing up 'server-level Linux distributions' is IMHO anybody's guess.

Share this post


Link to post
Share on other sites

There have been further contributions from other posters to the Ars Technica Linux Kung Fu forum thread I started on the technical feasibility of enforcing "In a future update, Linux clients running on server-level Linux distributions will be treated as server clients."  The last poster came up with the idea of a "combined-score version" of the Retrospect Client for Linux x64, but then said:

Quote

Not worth it for any software developer to spend a lot of time/money on anti-circumvention technology for their products on Linux, because of all the major platforms, Linux users are the most skilled, knowledgeable and resourceful. I was just talking about hypotheticals. The developer could hypothetically check, but they probably won't. There's a lot of software out there that's licensed in different ways for different users/machine configurations, and the only "enforcement" is on honour basis. For example, Teamviewer asks when installing you if you are using it as a home or commercial user, and will ask for a license key for the latter, but they don't actually audit you to see that you are a home user. They just trust you that when you check "Home", you're not going to use it in a commercial environment. I'd say MOST software that's dual licensed, or has different tiers of licensing/pricing (based on number of seats, for example) is probably enforced on an honour system, in fact.

The reason that Microsoft actually goes out and physically audits companies for license compliance is that Microsoft software installation base is in the billions, so if a small number of enterprise users cheat the system (they only audit enterprise users, they don't really check small businesses, and definitely not for home users), they lose A LOT of money. Most software doesn't have such a large installation base.

I then said, in the single Ars Technica Macintoshian Achaia thread where I'm allowed to name Retrospect:

Quote

I conclude from those posts that it would be impractical for Retrospect Inc. to actually enforce "In a future update, Linux clients running on server-level Linux distributions will be treated as server clients." So all they could do actually is to either charge all customers using the Retrospect Client for Linux x64 a fee that I hope would be rather small, or to put all such customers on the honor system for paying a larger fee if their Retrospect Clients for Linux x64 are being used to back up large-scale servers.

 

Share this post


Link to post
Share on other sites
On 11/1/2018 at 1:27 AM, DavidHertzberg said:

DovidBenAvraham has now manged to do two things:

....

Second, based on the discovery by JamesOakley that "under 15.x for Windows, even Desktop edition gets multiple execution units,"  I was able to verify that the same is true for "activity threads" under Retrospect Mac 15.0—using the test version Retrospect Tech Support had given me for diagnostic logging of -530 bugs.  (I still run production backups using the Retrospect Mac 14.6 "backup server" and the 14.1 Client, because later versions disable my -530 workarounds.)  DBA has now managed to work an announcement of that feature into this section of the "Retrospect (software)" article.  The new feature, of course, is simply a disabling of the Desktop Edition code that previously reset the number of units/threads to 1 every time the Engine was started—thus forcing Desktop Edition to mimic the non-multithreaded Retrospect Windows/Mac 6.  :rolleyes:  Retrospect Inc. is apparently so reluctant to mention this improvement that DBA had to reference it with a roundabout bug fix note in the cumulative Retrospect Windows Release Notes. <_<

Retrospect Mac administrators,

When the Retrospect Inc. engineers corrected Retrospect Windows 15.1 so  that "Fixed: Desktop can now reduce its maximum execution units (#7317)", they didn't look at the Retrospect Mac 15 Console.  Running Retrospect Mac 15.6.1, if I set General->Preferences Allow n Activity Threads to set n lower than 16, it snaps back to 16 whenever I reboot my "backup server"—thus restarting the Engine.  However if I merely quit the Console and then restart it again, that preference stays where I set it.

I'll file a Support Case.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×