On Fri, Aug 03, 2012 at 09:25:15AM +0200, Michal Kubeček wrote:

On Thursday 02 of August 2012 14:21EN, Rajko wrote:

...

This is what new bug reporters and everyone else that needs Novell account have to fill in before they can access openSUSE infrastructure:

https://secure-www.novell.com/selfreg/jsp/createAccount.jsp

Don't you think that is a bit excessive if we want more contributors?

"Novell Customer Center" in the page title doesn't help much either...

There is a branded page for this, which is linked from the opensuse.org pages: https://secure-www.novell.com/selfreg/jsp/createOpenSuseAccount.jsp?%22https... Just not from bugzilla.novell.com, more or less obviously. Ciao, Marcus -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

Rajko wrote:

On Fri, 3 Aug 2012 09:52:02 +0200 Marcus Meissner <meissner@suse.de> wrote:

...

There is a branded page for this, which is linked from the opensuse.org pages:

==== This is a 1 time thing that needs to be done (I did it ~ 8 years ago)...

All of your examples were reduced to 2 clicks -- 1 on login, and 1 to enter my info... They could keep keep me logged in longer (don't expire the cookies so fast).. but given the "hassle" of 2 clicks (no key presses)... it's hard to complain ... so far my 1 login has mostly worked everywhere except the OBS system...(of course!, it hates me!...)... -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Tue, 07 Aug 2012 18:03:30 -0500, Rajko wrote:

On Tue, 07 Aug 2012 12:45:41 -0700 Linda Walsh <suse@tlinx.org> wrote:

...

This is a 1 time thing that needs to be done (I did it ~ 8 years ago)...

Right, but it is not about number of clicks to login. It is about information one has to provide to sign up, that some people consider excessive. They report bug, but no amount of time spent convincing them to sign up, changed their mind about filling bug report directly at bugzilla.novell.com .

It's not necessary, though, to provide accurate information. AFAIK there's not a lot of even field format validation on those forms. But perhaps the SUSE BU can find out if there's a way to require less information for SUSE-specific (or even openSUSE-specific) registrations.

It is about 3-4 openSUSE users that I know about and it is for period of last few years, which is insignificant, if one would not know: http://www.useit.com/alertbox/participation_inequality.html

Due to this issue alone, openSUSE lost quite some number of reports, and most likely users.

I don't see that users would be lost as a result of this - since there's no login requirement to download the ISOs nor is that used for ML access or IRC access.

...

... They could keep keep me logged in longer

Inactivity counter is about 4 hours.

For Access Manager, that is true. But for the web app itself, it may maintain its own session timer. There are sometimes weird issues (we see them in the forums occasionally because vBulletin has its own session timer as well) because "activity" is defined as "interacting with the Access Manager system" - so if you do something that would require a login, AM checks to see if you are authenticated already, and if you are and have access, you are just passed through - and AIUI the session timer is extended at that point. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Wed, 08 Aug 2012 09:20:40 +0200, Michal Kubeček wrote:

On Tuesday 07 of August 2012 18:03EN, Rajko wrote:

...

Inactivity counter is about 4 hours.

I don't think it is an _inactivity_ timer as it logs me out even if I do work with bugzilla.

It is and it isn't. But the definition of activity isn't "using a service that uses Access Manager for login", it's "accessing something that requires validation of the authentication credentials". You can work for hours in Bugzilla and never do anything that would force you to login if you weren't already logged in. The same thing happens in the forums. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On 2012/08/09 08:38 (GMT+0200) Michal Kubeček composed:

In general, yes. But almost all bugs I work with require login for any action, even to just view them. Nevertheless, I still have to log in at least twice a day (not counting the first login in the morning).

To be clear, I use many bug trackers, most of which use Bugzilla. Novell's is the *only* one that logs me out according to its own schedule. The rest at a minimum stay logged in at least until the browser session is closed. It's highly annoying, and I did file a bug about it: https://bugzilla.novell.com/show_bug.cgi?id=753203 -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

Rajko wrote:

On Tue, 07 Aug 2012 12:45:41 -0700 Linda Walsh <suse@tlinx.org> wrote:

...

This is a 1 time thing that needs to be done (I did it ~ 8 years ago)...

Right, but it is not about number of clicks to login. It is about information one has to provide to sign up, that some people consider excessive. They report bug, but no amount of time spent convincing them to sign up, changed their mind about filling bug report directly at bugzilla.novell.com .

It is about 3-4 openSUSE users that I know about and it is for period of last few years, which is insignificant, if one would not know: http://www.useit.com/alertbox/participation_inequality.html

Due to this issue alone, openSUSE lost quite some number of reports, and most likely users.

I completely agree with the core issue here, but I suspect we have more unfixed/largely ignored reports than we have lost reports due to this. :-( -- Per Jessen, Zürich (19.8°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Wed, 8 Aug 2012 17:10:47 +0000 (UTC) Jim Henderson <hendersj@gmail.com> wrote:

... but without concrete data, it's difficult to even prioritize this possible issue.

Core issue is that we have mess in login, sign in, sign up, register path, and it is not possible issue. If you have mess on a first step that new user has to take, then what one can expect later. It is almost funny that with so many people around no one complained about "Sign up" link that leads to another Login (Sign in) screen not to the Register (Sign up) as one would expect. Of course it is small difference between "Sign in" and "Sign up" that non-native English speakers can miss, but where were all those that have no such problem? It is hilarious that people come with indirect advice to fake data in a form (data integrity is not checked), in order to satisfy customer registration form, which is not intended for creation of openSUSE account, but Novell customer account, but it is sad that after 4 days detailed report I checked again and nothing has changed in the issue that requires nothing more then to fix few links. Who should know about this to expect fix in another few days? (Let me guess, bugzilla.novell.com :) -- Regards, Rajko -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On 08/09/2012 04:42 AM, Jim Henderson wrote:

Significantly less than 1% of those people have complained about the process it took to get registered - and the openSUSE forums have always used this registration process since their inception with the forum merge project.

the people who refused to provide all the personal data (*or* enter fake data) were not likely to press on and complete the process in order to be able to complain.. instead, i propose they just closed the form...and left.. or, maybe they hung around, read the forums/wiki and helped themselves... or, used nntp to completely avoid both the sign-up and blinkies.. or, maybe they found any of dozens (hundreds?) of other online fora where Linux and openSUSE questions are fielded.. or, maybe they just went back to Windows, Ubuntu or wherever they were coming from...along with a "bad feeling" like: Why oh why would openSUSE want all that info!?! i mean afair (and its been a LONG time) before i could participate in their Q&As none of Microsoft, IBM, Red Hat or etc etc etc wanted to know _my_ street address, job title, and *telephone* number before i even asked the question!! imho the entire form <https://secure-www.novell.com/selfreg/jsp/createAccount.jsp> is a huge turn off for most anyone likely to give Linux a try.. in any case the rationale against changing/improving because of "almost no one complained" is very weak at its strongest.. and the point that "concrete data to work with" is needed: please ask Novell to open up their records since the fora were created and provide you/us with the number of hits over time that that page got hit, but no new registration resulted.. ymmv dd -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On 08/09/2012 05:52 PM, Jim Henderson wrote:

On Thu, 09 Aug 2012 07:34:48 +0200, DenverD wrote:

...

imho the entire form <https://secure-www.novell.com/selfreg/jsp/createAccount.jsp> is a huge turn off for most anyone likely to give Linux a try..

My point is that this is an opinion, not a set of data showing how many times people got to that page and didn't complete filling it out. "Gut instinct" isn't a very good measure of whether something needs to be done.

...

in any case the rationale against changing/improving because of "almost no one complained" is very weak at its strongest..

I would disagree. I've spent years making decisions based on data driven analytics rather than gut instinct.

if the data you demand exists, it is held by Novell.. will they provide it to the community? dd -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Fri, 10 Aug 2012 18:02:21 -0500, Rajko wrote:

...

if the data you demand exists, it is held by Novell.. will they provide it to the community?

No.

When I was serious about wiki I needed data, and I would be happy with some portion of server logs, but privacy and other stuff made impossible to have even portion that will contain only page access.

Instead offer was to see some analytics for particular pages, and not too many of them as it was time consuming to generate reports. As my goal was to identify pages that need attention, offer was like catch 22, not very encouraging.

Arguably, that was then, this is now. Now we have Agustin and his team, who may be in a position to get this data internally and use it to inform project decisions without compromising the corporate obligations to the users registered in the system. But like I said to DenverD, I've already made the request through people I know who are involved on the "overall" web team (managing the corporate sites and the analytics data), so perhaps something can be worked out. I'm happy to make the effort so we aren't guessing about the effects of the registration process and trying to fix either a problem that isn't significant (in terms of possible attrition/loss of users due to the registration process) or the wrong problem altogether. Unfortunately, they switched from Omniture to Google Analytics a couple years back, so I'm not sure how much detail is available as I've never use GA myself for following click paths through the site. Omniture was really good for that kind of analysis. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Fri, 10 Aug 2012 23:09:08 +0000, Jim Henderson wrote:

...

if the data you demand exists, it is held by Novell..

It's held by The Attachmate Group and SUSE (in a shared data store shared with Novell).

...

will they provide it to the community?

I might be able to obtain it. I'm asking around to see if I can get it, even in a generic form that lets us see the percentages of registrations abandoned and completed.

One of the challenges is going to be that not all the openSUSE pages are Google-Analytics enabled yet, it seems - and that the path goes from opensuse.org to novell.com (for the login page) to suse.com (for a redirect to the account creation page) and then back to secure- www.novell.com. GA doesn't seem to handle cross-domain paths very efficiently when following paths. But progress is being made on trying to find this out. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Thu, 09 Aug 2012 07:34:48 +0200, DenverD wrote:

...

and the point that "concrete data to work with" is needed: please ask Novell to open up their records since the fora were created and provide you/us with the number of hits over time that that page got hit, but no new registration resulted..

Forum registrations are actually something that I did an analysis of a couple months ago - based on the first date/time users posted and the last date/time users posted on.

Linus making his noise about openSUSE's security being a PITA caused a larger dip in traffic and new registrations than anything else over the course of the forums.

Traffic patterns in the forums compared to the mailing lists (where no registration through "Customer Center" - which is where the information is effectively stored) are surprisingly consistent, thought the volume across all of the support-oriented forums in OSF is much higher than the traffic on the opensuse-users mailing list (I only looked at the English mailing list).

There are definite spikes in traffic that lag each release slightly. IIRC, there was a discussion of the traffic analysis I did back in May on this very mailing list (though I may be misremembering). I even ended up inadvertently having my access through gmane blocked as a result because I used their NNTP interface to the ML to suck the headers out (not even considering this would incorrectly identify me as a possible e-mail harvester/spammer - odd in itself since e-mail addresses are munged anyways).

What the analysis showed was that both the MLs and the forums have had a declining trend in traffic overall. The two trends are in similar fashion over time.

The forums ramped up (with some spikiness) from June 2008 through November 2009, and then trend downwards at a somewhat steady rate.

I see similar trends on the opensuse-user mailing list in posting trends. The ML history that I pulled went back to 2006, but aligning the data to the same date, the ML traffic is much smaller (about 2000 messages per month average over the life of the mail lists vs. ~8000/ month on the forums per month over the life of the forums - both through May 20 when I pulled the information).

What this analysis told me then (and tells me now looking at it with fresh eyes) is that the reason for the decline in user participation isn't primarily driven by the registration process.

To address the participation problem, based on the data I've looked at myself, we need to identify what is reducing participation. The registration requirement does not appear to be the cause.

Looking at it from a raw numbers perspective, the forums have a slightly steeper downward trend, but evaluating based on percentage of total message traffic in each venue per month (both measured from the date the forums opened), the trend for the mailing lists is alarmingly steep (but that may be because the forums opened up a new venue so people who used to be on the MLs moved to the forums - possibly). Looking at it over the full dataset for each venue, the mailing lists decline percentage-wise slightly faster.

The bottom line, though, is that we have a decline in user participation in both venues - and that there's a factor that's common to venues that require registration through customer center and venues that don't. We need to identify *that* cause and address it.

How do we do that? Maybe by finding out who hasn't been around and asking them why. Do a survey and ask - is it just because you like trying different distributions? Is it because you used to and landed somewhere you were happy - and it's nothing to do with the openSUSE project itself, or was it because of something specific that happened with openSUSE? Or was it because Linus slammed openSUSE for security back in March? Or maybe it was because of the significant changes in KDE and GNOME and a more lightweight desktop as default is desired? (Debian recently announced they were switching the default DE to XFCE; maybe they know something we don't?)

What could we do to bring those users back? Maybe asking them would be a good way to find out.

Seems that's a better approach than making assumptions and trying to fix a problem that might not even be a primary cause that people left.

Jim Hi Some of it would/could also mean that openSUSE overall is improving and users are having less issues to require forum/ML/IRC

On Sat, 11 Aug 2012 00:16:23 +0000 (UTC) Jim Henderson <hendersj@gmail.com> wrote: participation? -- Cheers Malcolm °¿° (Linux Counter #276890) SUSE Linux Enterprise Desktop 11 (x86_64) Kernel 3.0.34-0.7-default up 12 days 21:58, 3 users, load average: 0.90, 0.42, 0.42 CPU Intel i5 CPU M520@2.40GHz | Intel Arrandale GPU -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Fri, 10 Aug 2012 20:42:46 -0500, Malcolm wrote:

Hi Some of it would/could also mean that openSUSE overall is improving and users are having less issues to require forum/ML/IRC participation?

That certainly is a possibility. It would be good if there were other things we could measure (downloads per release, perhaps?) to see what the trend is for at people at least wanting to have a look. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Fri, 10 Aug 2012 23:43:26 -0500, Malcolm wrote:

Hi Gives an idea for OBS; http://www.suse.de/~coolo/repo.list

Do you know what the numbers in the left-hand column represent? Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Sat, 11 Aug 2012 10:03:27 +0200, Andreas Jaeger wrote:

On 08/11/2012 07:15 AM, Jim Henderson wrote:

...

On Fri, 10 Aug 2012 23:43:26 -0500, Malcolm wrote:

...

Hi Gives an idea for OBS; http://www.suse.de/~coolo/repo.list

Do you know what the numbers in the left-hand column represent?

I guess number of packages downloaded - or number of repo accesses.

In either case, it would then be useful to know over what time period. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Thu, 09 Aug 2012 09:35:21 +0200, David Majda wrote:

The fact that every additional field in a form lowers the percentage of people that complete it is a well established fact by multiple studies. Just Google e.g. for "conversion rate number of form fields" and spend some time going through the links to see some examples. Or read any good usability book.

I consider the above to be "hard" enough. I don't see any reason why the registration form should be an exception.

I would far prefer to see analytics that show the number of times someone visits the registration page via an openSUSE site (ideally) and doesn't complete the form. That would provide specific data as to the number of times the form was reached and wasn't completed - for whatever the reason. One might argue that given the nature of the registration form - that Novell and SUSE uses for the registration are typically business- orientated, which would mean you won't have hobbyists (generally) would have high rates of completion - so the number of abandoned visits to the page would still be a reasonably accurate measure for openSUSE, where there's not a business driver on the end of the user filling the form out (which is an instance where the option exists to say "screw this" and either move to another distribution or just opt to not participate.).

...

That's not to say there isn't a problem, but as a friend of mine who works for Google is fond of saying, "the plural of 'anecdote' is not 'data'."

The only reliable way to see how the complex and personal-data-acquiring form affects the number of registered users I know of is an A/B test: prepare a simplified version of the form and measure the difference in registration rates.

As a predictive measure, yes. As a historical measure, web analytics can provide some useful insights.

...

On the forums alone, we have currently 69,105 registered members - which means we have that many people who have gone through the registration process for the forums alone - and once that's done, they have access to Bugzilla, the wiki, and everything else in the openSUSE project that requires authentication. Even SUSE Studio and OBS.

Significantly less than 1% of those people have complained about the process it took to get registered - and the openSUSE forums have always used this registration process since their inception with the forum merge project.

People who don't complete the registration process don't complain, they leave.

Three classifications exist: 1. People complete and don't complain. 2. People complete and complain. 3. People leave and don't complain. It's *usually* (but not always) a good indicator if you have a significant number of people in category #2. If someone found it cumbersome but felt that the value existed in filling the form out anyways, they'll usually complain. 1% of the registrations that relate to the forums (which is a subset of all registrations related to the openSUSE Project) would be about 600 complaints. If there have been about 700 complaints, I'm pretty blind because I simply don't remember seeing them. I remember seeing maybe half a dozen people ask "what's up with all this information", and most of those people have been on the mailing lists that I read.

Also, the fact that "it was always like that and worked somehow" is not a relevant argument here. Computers also worked well when they had 640k RAM but they have GBs today. The question is whether this can work better. Maybe these 69k users could have been 138k today if the form was simplified long time ago.

Sure. And I'm not nor have never said that we *shouldn't* do this. I'm saying in terms of setting priorities, identifying the severity of the issue is significant. In project management, you don't spend a lot of time or effort on something that's impacting you by 10%. You spend it on things that are impacting you 80% or more. When you've dealt with the most severe problems, then you have the luxury of looking at the smaller issues. Otherwise you're working on the wrong things.

...

Note that I'm not saying it's perfect by any stretch. I'm just saying that relative to the other issues that have been under investigation, a working (if even somewhat convoluted) registration process is relatively "small potatoes".

To me, it seems weird to question the priority of this in 2012, when every startup optimizes their front page and registration process like crazy, people often refuse to buy in e-shops requiring registration, and there is an entry on Hacker News about a workflow simplification leading to increased conversion rate every month.

Again that's still anecdotal evidence, though. Just because everyone is spending their time doing something or following the latest "hot trend" doesn't mean we should. To reiterate, though - I'm not saying "we shouldn't spend time on this", I'm saying "let's find out if we really need to spend time on this". For my part, I can see if I can actually obtain some data that says we've got people who are not completing the registration process and how big of an issue it really is. It may be a bigger issue than I think it is, in which case I'd fully support efforts to reduce the number of people abandoning the project before they even get started with it. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Thu, 09 Aug 2012 19:03:13 +0200, Carlos E. R. wrote:

You can have a look at the registration data people write and see how much is faked. I suppose Novell keeps that under 7 keys, but I get a glimpse of that by looking at the login names in the forums, with impossible random combination of letters, or logins that intentionally point to why do you care for my name. It would be interesting to look at that data: it would be hard data at how much people despise the registration process.

That's not really a good measure of anything other than how many people submitted bogus information. Doing analysis of all the variations of "why do you care?" would be incredibly time consuming and would provide little value compared to looking at how many people just stop at that point in the process. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Thu, 09 Aug 2012 12:28:26 -0500, Bryen M Yunashko wrote:

I believe the heart of the frustration for why this topic keeps coming up over the years is not about what's right or wrong with the infrastructure, but that the community itself doesn't really have any control over the infrastructure. In our openSUSE community, we're a community of "just do it"-ers. People see something that they feel needs to be worked on or is interesting enough to them, they can just step up and do it.

That makes sense, and yes, there is an element of "we can't touch that part of the infrastructure". I do have regular contact with those who do, so I do what I can - but certainly in terms of changing the registration form or something like that - no, that's not something we would have access to.

Obviously, they cannot do that when it comes to the infrastructure that we live in here. We are unable to assign work to handle certain points of our infrastructure. We are unable to utilize the many and varied expertise within the community. We can't even say "Well fine, you're gonna complain? Here, you fix it!" Can we ever get to a point where we can do that, with existing infrastructure or new infrastructure? I don't know. But until somehow community feels it has a measure of control, influence and sense of ownership over its infrastructure, this topic will always keep cropping up.

Absolutely. But identifying if there's a problem vs. the convenience (and it certainly would be a HUGE convenience for the community to manage it) and weighing those factors is something that does need to be identified. Classic benefit analysis would apply to that. If the bottom line is control, then let's talk about it in terms of control rather than in terms of an unmeasured/possibly unmeasurable number of people who gave up before they got started. Convenience and control is something that's measurable. But having a relatively small group of people who regularly bring it up very loudly doesn't increase the size or scope of the issue. It's not a "loudness" debate, nor does it have to be. Maybe someone like Alan or the Boosters team could help us here in terms of working with the internal team that manages this registration process to create a second path that asks for a very minimal amount of information. I don't see that exporting the user data into our own datastore is likely for legal reasons (since there is corporate data as well as openSUSE user data in the datastore itself) - but I could be wrong about that. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On 08/11/2012 01:10 PM, Rajko wrote:

On Thu, 09 Aug 2012 12:28:26 -0500 Bryen M Yunashko <suserocks@bryen.com> wrote:

...

But until somehow community feels it has a measure of control, influence and sense of ownership over its infrastructure, this topic will always keep cropping up.

That is part of another bigger problem.

We discuss here issue that affect us, but that we can't fix, yet, no one from SUSE management, that is involved in that part of infrastructure, posted even "I'm busy right now, but see you in a week".

What message sends that lack of interest? IMHO, that me and everyone else discussing this can use time somewhere else, instead of wasting it here.

Rajko, I'm sorry to see your frustration. Agustin and the openSUSE team would need to look into this and Agustin send out yesterday an email saying what their current priorities are. Let me copy Agustin explicitely on this email to make sure that he reads the whole thread and learns about the problems of our infrastructure in general and account creation specifically, Andreas -- Andreas Jaeger aj@{suse.com,opensuse.org} Twitter/Identica: jaegerandi SUSE LINUX Products GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Jeff Hawn,Jennifer Guild,Felix Imendörffer,HRB16746 (AG Nürnberg) GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126 -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Tue, 14 Aug 2012 12:41:21 +0200, Jos Poortvliet wrote:

On a personal note: I have gone through this list of questions before I joined the company and I agree it's annoying to say the least. If there's a relatively low-effort way to fix it, that'd be awesome. If it's a lot of work to fix it I agree with Jim: we need some numbers first. In both cases I wouldn't know where to start but again, if you like, ask formally and we'll try to figure it out.

What I've been able to find out so far is that there's not a lot of data available - many/most of the openSUSE sites are lacking the necessary javascript to hook into Google Analytics so we can get that data. I did raise a bug on the redirect page that comes up when trying to create a new account starting from an openSUSE site - it's bug 776082. The redirect auto-refreshes, but it's a delay that can be removed. There also is a "short form" registration page that we can look into using for account creation. It'd need to be re-skinned but otherwise it should serve the purpose and be an easy thing to implement. Maybe this should be a topic on the next openSUSE project meeting? Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Thu, 16 Aug 2012 17:38:36 +0200, Andreas Jaeger wrote:

On 08/16/2012 05:32 PM, Jim Henderson wrote:

...

On Tue, 14 Aug 2012 12:41:21 +0200, Jos Poortvliet wrote:

...

On a personal note: I have gone through this list of questions before I joined the company and I agree it's annoying to say the least. If there's a relatively low-effort way to fix it, that'd be awesome. If it's a lot of work to fix it I agree with Jim: we need some numbers first. In both cases I wouldn't know where to start but again, if you like, ask formally and we'll try to figure it out.

What I've been able to find out so far is that there's not a lot of data available - many/most of the openSUSE sites are lacking the necessary javascript to hook into Google Analytics so we can get that data.

openSUSE is not using Google Analytics since some time, it's using Piwik,

Thanks, Andreas - I wasn't aware of that. Unfortunately, that makes it difficult to track back to the Google Analytics stuff used on the login/ registration pages, since those pages are tracked from GA. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Tue, 14 Aug 2012 12:41:21 +0200 Jos Poortvliet <jos@opensuse.org> wrote:

I don't understand where this comes from.

It comes from frustration Jos, as AJ said, and his answer was the right one. At least some of SUSE managers read this, Board should read this, so it is reasonable to expect that issue is escalated to right insiders without formalities and it is reasonable to expect at least some response after a week of discussions that obviously lead nowhere without right people involved. It'll be good if answer would give some directions where to look for a solution, but telling that one has no time is an answer, much better then no answer at all.

... if you like, ask formally and we'll try to figure it out.

See above :) My experience with account creation page is mentioned in couple of emails. Annoyance is not only amount of data asked, but also bugs in the form(s) (so far: wrong links, ineffective check box, multiple forms with the same purpose, too many clicks involved) -- Regards, Rajko -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Thu, 09 Aug 2012 09:35:21 +0200 David Majda <dmajda@suse.cz> wrote:

Just Google e.g. for "conversion rate number of form fields" and spend some time going through the links to see some examples. Or read any good usability book.

Thanks for the pointer. Usability http://useit.com . Following this one will be sufficient: http://www.quicksprout.com/2012/06/25/5-ways-to-improve-your-contact-form-co... About "conversion rate" http://en.wikipedia.org/wiki/Conversion_rate There were some stats that reminded me that one can't trust diagram created by unknown methodology: http://blog.hubspot.com/blog/tabid/6307/bid/6746/Which-Types-of-Form-Fields-... 40,000 pages must be very different. Forms can serve very different purposes and there is no way that one can make meaningful statistic out of that. On e-commerce site payment info form can have 10 fields and anyone buying merchandise will not run away. Conversion rate losses close to 0%, but if that would be login form, loss will be closer to 100%. As mentioned in quicksprout.com link, it is all in timing. Ask upfront and you lose customer, ask later when is obvious that you need that info, and no complains. -- Regards, Rajko -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Sat, 11 Aug 2012 05:34:29 -0500, Rajko wrote:

On Thu, 9 Aug 2012 02:42:10 +0000 (UTC) Jim Henderson <hendersj@gmail.com> wrote:

...

...

But Rajko, that's still anecdotal, not hard data. It's guessing there must be a problem because of how it's designed.

It is not that much guessing.

But it is guessing that we're losing people. If we could look at how many times the page is actually abandoned rather than completed, that's not a guess.

There are principles how to design human interface, which are developed by the people that understand how human body works. After visiting a lot of web pages, and reading a lot on topics described in http://useit.com I tend to guess much lesser then those that didn't.

In other words when you see roof under the house, and tell that is wrong, is that a guess that there is a problem, just because how it is designed :)

Sure, but if people still are completing the form, then it's not as big of an issue as other things and can be prioritized lower.

On the other hand, when you see http://www.yproxy.com/blog/quality-control-on-name-brand-electronics/ do you see a problem instantly? Maybe.

I do. It is what I was busy with for years, and having an eye for bad solder made my life much easier. There are details that will not attract attention of casual viewer, but will be considered important by informed person.

Sure, but that's something where you can see the immediate result of the effect. Yes, it looks like poor solder joints, but the thing that prompted the individual to look was an actual measurable problem. If the picture hadn't given out, he wouldn't even have gone looking.

Back, to registration and login pages. In this case I'm just informed viewer. Professional web and UI designer will find far more mistakes.

Anyone with professional training can find things they'd improve in something they look at. That's generally a matter of an informed opinion, but I've known professionals who had deep knowledge of matters who have had different opinions about how to proceed.

...

That's not to say there isn't a problem, but as a friend of mine who works for Google is fond of saying, "the plural of 'anecdote' is not 'data'."

And what your friend at Google tells about this tower: http://en.wikipedia.org/wiki/Leaning_Tower_of_Pisa When you see that building, do you think for a second that is a good design? It is hundreds years old and it still stands, like our registration process. Not that anyone sane will climb up, but it stands.

Except that we've got a large number of people who have "climbed up" and the vast majority haven't complained about it.

...

On the forums alone, we have currently 69,105 registered members - which means we have that many people who have gone through the registration process for the forums alone - and once that's done, they have access to Bugzilla, the wiki, and everything else in the openSUSE project that requires authentication. Even SUSE Studio and OBS.

All fine. Try to read this again: http://lists.opensuse.org/opensuse-project/2012-08/msg00014.html it is not the same experience on different sites. Some have no chance to see openSUSE registration page.

There again, matter of opinion, though. If we perfect the registration process to the anecdotal standards desired, we'll still be able to go to other sites and say "hey, our experience isn't the same as on different sites". On the forums, for example, people complain about not being able to edit their posts for an indefinite period of time. Others point out that the reason for this because we gate the messages to NNTP, which is something else that other sites don't do - but we do to provide an additional access method for those who prefer that interface.

...

Significantly less than 1% of those people have complained about the process it took to get registered - and the openSUSE forums have always used this registration process since their inception with the forum merge project.

Check again. It is not the same. Now forums have access to shorter openSUSE registration form. Also, check number of registration before and after switch.

What "switch" are you talking about? If I have about 70,000 people who have registered through the existing registration system and only a few (I think maybe 3 or 4 total in the forums themselves) have complained, that's statistically a very small number who have complained about the issue.

...

...

...

Who should know about this to expect fix in another few days? (Let me guess, bugzilla.novell.com :)

While there are still issues being fought with regards to login performance and whatnot, I would expect the prioritization of something that has generally been seen to work well (or at least not cause a lot of complaints for all those who have gone through the process).

First I still need answer on who should know? It seems that all that can bring problem to solution don't read this list.

I would start with the openSUSE team at SUSE.

Then me and complains.

Did I complained when I went trough? No.

How many times I visited registration page before I finally decided to go trough? At least twice asking myself:"Why in the hell they need so much?" and left.

How many people we lose each day that will be good contributors, but never manage to go tough? No one can tell. Those that go away don't talk.

Right, but those who complete the process but were unhappy with it are very likely to complain. I've done this myself with some things - go ahead and complete the transaction (whatever it is), and then provide feedback saying "yes, I got in, but the process was very convoluted/ onerous/etc".

What is impact of one person not signing up? IMHO, big. More people will contribute more, improve quality and attract more other contributors, like avalanche.

...

Note that I'm not saying it's perfect by any stretch. I'm just saying that relative to the other issues that have been under investigation, a working (if even somewhat convoluted) registration process is relatively "small potatoes".

I guess that you did project management and put small easy to fix issues ahead of bigger that need much more time to fix. There is always end of the day when it is not reasonable to start big tasks, but too much time to chat.

Yes, I did project management - and sometimes quick fixes can be put in place quickly. But at the same time, (as I say below), it's one thing to assume how much work is involved, it's another thing to know how much is actually involved based on how the systems work. Over the years, I've learned that things are rarely as simple as they look.

This one is really about 1 hour of work for skilled person, if we don't count administration overhead. Keeping that unresolved until more pressing stuff is addressed will keep small issue(s) working against us.

For someone who doesn't know how a system is implemented, works, or what the requirements were at the time of implementation, sure, it can look like it's only an hour of work. The reality of that is often (not always) different.

...

I wouldn't expect it to be "fixed" in a short period of time because it is at least working, if not what some (myself included) would call "efficient".

It is more then efficiency. It favors people that don't think twice before they jump.

Which is something that should be measurable - and once we have a measurement, we can know if it's worth spending time on.

...

Jim

-- Regards, Rajko

-- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

Rajko wrote:

On Tue, 07 Aug 2012 12:45:41 -0700 It is about 3-4 openSUSE users that I know about and it is for period of last few years, which is insignificant, if one would not know: http://www.useit.com/alertbox/participation_inequality.html

Due to this issue alone, openSUSE lost quite some number of reports, and most likely users.

OpenSuse Doesn't want those reports. If they can't even follow the corporate line in regards to creating a login, then how are they going to follow open-suse mindthink that further narrows the fields of what is even ***ALLOWED*** to be reported as a bug without being abused? See https://bugzilla.novell.com/show_bug.cgi?id=771516 For a great example of something that used to work, but no longer does -- except under some ideal suse-build environment. It's a real bug report for something that used to work -- bug it now broken, yet the pressure (as documented in the report) has been to tell the user that it is their problem and that their valid, current, suse install the problem -- and is unsupported for a build environment. I had to fight just to have the bug closed as "WONTFIX" -- the developers wanted to close it as INVALID because their build works under their idealized build & test conditions. It used to be products were build to run on released products -- not just factory-only sterile conditions, but this was before openSUSE's Tivoization efforts to turn openSUSE into a platform for platform for developing end-user appliances. It's a retreat from the general linux-distro model of 1 source working anywhere (including other OS's) to an opposite and antithetical extreme of it working on virtually no end-user's real "in-use", production system (clean room build and test only!). This was most of my latest reply on that bug -- as it well details they problem(s) in the in the attitude shift that's caused the problem. The fact that most users are ok with this this (as indicated by the responder's last comment "this is how the community feels about your bug reports)" is a very strong reason why most users wouldn't be comfortable submitting feedback or bugreports. ----------------------- (In reply to comment #11)

If it is faulty please provide a patch to enhance the working version. ==== The purpose of a bug-tracking mechanism is not to force nor beat users into submission for having submitted a bug. It's not to prove the superiority of your build process over theirs. It's not to prove to them that your idealized value of how to run the program works and whatever they are doing is wrong.

The point is to get feed back about something that broke for the user -- specifically something that *USED* to work (building samba from RPM) for them. This is a regression. It used to be those were treated as P1/S1 bugs. Now the user is abused for not having an approved setup and the user has to try to explain why they bug report isn't invalid.

It's your build which fails and I don't have the time nor do I see the value to the project to debug your failing build.

At one point in time people cared about the projects and products they shipped on to build in as many configurations (to the point that they build on DIFFERENT OS's). You only care that your build works on some idealized config -- which is very different from the attitude of how open source development used to be done.

It is faulty for your requirements but not for the requirements we have as an Open Source Software project.

The Samba project builds on multiple architectures with no were near the restrictions of the open suse project. Opensuse's version of rpm build is faulty because it doesn't build -- NOT just in all the places that samba builds, but because it doesn't even build on "generic" (not specially tuned) machines as does the original script. Why? because you remove various pieces from samba necessary for it to build all of itself. I cannot submit a patch -- because it would be rejected. I have been that route before -- It's not because I can't submit a patch -- it's because suse only wants certain things two work. If you want to swear that you will accept my patch, I can create one -- but I would bet I'll hear that "users don't want all that other stuff on their machines (ldb, tdb, docs...etc)... You are creating 'use-only' machines by design (that don't easily support modification or developers). I can't create a patch for that as it's a high level decision made about products more than just samba. It's what stallman calls the 'Tivo-ization' of the product.

I don't care if the bug state stays at wontfix. It demonstrates very well your unwilling habit to learn and cooperate.

---- Only if I'd never submitted patches that had been rejected would that be true. But having done so, because it's not the direction open suse has decided to take, the bug staying at won't fix is documentation of configurations that suse doesn't support. It isn't a reflection of your development abilities. It documentation of an area that Suse no longer supports. This information is vital if open suse is ever to figure out how their decisions affect those who try to use all of suse as a development platform and the effects of their Tivoization program in moving openSuse to be an appliance development platform rather than a linux development platform.

See also comment #1 to get a picture what our community thinks.

It shows me very well the picture of a user that says "I got mine", and the fact that your build, which used to work, no longer works is your problem, and unless you apologize for wasting our time for having even submitted a case were something doesn't work, we are gonna ignore you for having a system we no longer support. Do you really think their weight counts any more than me than that of a petulant 5-year old? This note is a perfect example of why it's pointless to worry about people registering for bugs -- when they will get this type of response to reporting them. It's called "hypocrisy". -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

Jim Henderson wrote:

On Thu, 02 Aug 2012 14:21:42 -0500, Rajko wrote:

...

This is what new bug reporters and everyone else that needs Novell account have to fill in before they can access openSUSE infrastructure:

https://secure-www.novell.com/selfreg/jsp/createAccount.jsp

Don't you think that is a bit excessive if we want more contributors? Is there any other way to have account?

I did some asking around, and it looks like there is a simple registration page that could be leveraged for this - it'd need to be reskinned, but it looks like the bulk of the work is already done.

https://secure-www.novell.com/selfreg/jsp/createSimpleAccount.jsp is the Novell version of the "simple account" page.

Apart from the theming/skinning, how does this look?

Looks good to me. A nice-to-have to would be auto-selection of country based on IP. -- Per Jessen, Zürich (25.1°C) -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Fri, 17 Aug 2012 19:07:08 -0500, Rajko wrote:

On Mon, 13 Aug 2012 16:50:03 +0000 (UTC) Jim Henderson <hendersj@gmail.com> wrote:

...

https://secure-www.novell.com/selfreg/jsp/createSimpleAccount.jsp is the Novell version of the "simple account" page.

Re-skinning will not help much, it has to be rewritten.

Left sidebar is all Novell:

IIRC, the left sidebar is brought in with the CSS; as such, that's essentially re-skinning work. Relative to designing a completely new page and ensuring that the mandatory information needed on the backend is in place, adjusting this is likely a trivial thing.

Basic Novell Login Information: Make only email info mandatory and then drop "I am not associated with a company". It will make form simpler and more acceptable to plain users.

Perhaps a default of "not associated"? Some people may well want to specify a company.

Security question and answer: I don't think that security question and answer are necessary. It is used to recreate account access, where email should be better choice.

Security Q/A will give a chance to people that watch their passwords to make a mistake choosing simple word, or little known fact from their life, and create a backdoor with weaker lock then the main one. Of course, it will ask malicious side to do some research on the subject, but that is why it can be considered as a weak password.

These two are necessary - part of the integrated way in which account access is regained if a password is lost. Changing that has implications on the backend from what I understand. Remember that this isn't (nor is the intention to) completely rip out the backend - we have too much that depends on it. The goal is to simplify things for oS users. The implications of having to create a new account if a password is lost or stolen are pretty huge. Edits on the wiki, posts in the forums, access to SUSE Studio appliances, information stored in Connect - all of that becomes inaccessible. Better to have a way for the user to recover their account. (Not that that's really an option /not/ to have here - AFAIK it isn't) Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org

On Sun, 19 Aug 2012 13:50:11 -0500, Rajko wrote:

On Sat, 18 Aug 2012 05:55:48 +0000 (UTC) Jim Henderson <hendersj@gmail.com> wrote:

...

On Fri, 17 Aug 2012 19:07:08 -0500, Rajko wrote:

...

On Mon, 13 Aug 2012 16:50:03 +0000 (UTC) Jim Henderson <hendersj@gmail.com> wrote:

...

https://secure-www.novell.com/selfreg/jsp/createSimpleAccount.jsp is the Novell version of the "simple account" page.

Actually, BNC, http://bugzilla.novell.com has only one link to login page:

https://login.novell.com/nidp/idff/sso? id=5&sid=2&option=credential&sid=2

that is used by: Novell customers, SUSE customers and openSUSE users.

Whole change would be to add link to openSUSE registration: https://secure-www.novell.com/selfreg/jsp/createOpenSuseAccount.jsp

and of course to cleanup links on that page.

Yes, to customize bugzilla specifically for openSUSE would be a bigger issue because bugzilla itself is used by multiple businesses+openSUSE. Jim -- Jim Henderson Please keep on-topic replies on the list so everyone benefits -- To unsubscribe, e-mail: opensuse-project+unsubscribe@opensuse.org To contact the owner, email: opensuse-project+owner@opensuse.org