- 3,702,034 Users
- 2,239,551 Discussions
- 7,835,724 Comments
Forum Stats
Discussions
Categories
- 69 Community Feedback - NEW! (No Product Questions)
- 33 General Community Platform Concerns/Kudos/Feedback
- 32 Community Platform Bug Reports
- 42 How Do I? (Community Platform "How to" Questions)
- 10 Where is the...? (Community Platform Locations)
- 9 I'd Like to See... (Community Platform Ideas & Suggestions)
- 11.2K Community Feedback (No Product Questions) Archive
- 63 Personal Document & Blog Archive
- 2 Community Programs
- 2 Get-Togethers
- 46 Certification Community
- 4.7K Certification Community Discussions
- 31 Oracle Certified Master Profiles
- 25 Oracle Database 12c Administrator Certified Master Profiles
- 24 Visual Builder Cloud Service
Transition Feedback Thread
This discussion has been closed.
Comments
Thanks @Timo Hahn and @BluShadow - makes sense.
Subsequent mobile access worked fine, but I used URLs send by notification e-mails.
I think we need two ways to format code.
A) Real code with syntax highlighting in the correct (choosen) language.
B) non formatted monospaced text
A) would be used for larger code blocks. B) would be used for smaller snippets and to point out certain aspects.
Fortunately HTML already supports this using the <code> tag for a) and the <pre> tag for b). The new forum unfortunatly treats both html tag identical and applies non-working autolanguage detection upon them.
As a first step the syntax highlighting for the PRE tag should be removed. This would partially allow us to write code in the B) way.
In my first example the UPDATE statement was recognized as BASH probably because the second line started with SET, and SET is a bash command. This just shows how impossible the task of correctly guessing the code language is. The choosen syntax highlighter highlight.js is very well able to properly format code (see SQL example). The old forum used codemirror.js which is quite solid, but in the forum was a bit clanky to work with.
I'm extremely disappointed that for such a technical forum nobody seemed to have tested this important part before the forum went live.
I'm reading but staying out of this. But please keep going. Once the discussion does die down, I am taking the entire code thread to the developers so they can read through it.
Best,
Jim Finch
Community Experience Manager
System Admin
I find it so much slower than the old websites.
Is it me who finds that or is it a software/hardware problem.
I find it to slow and really unusable....
I've reported that to the developers myself. Though the more you visit the same areas, the faster it seems to get.
Best,
Jim Finch
Community Experience Manager
System Admin
Yeah but we cant rely on using browser caching for speed. 🙄
Though it is a work around for now, until the devs can do something about it. Something out of my control...
Best,
Jim Finch
Community Experience Manager
System Admin
Not sure if it is browser caching. It seems browser tries to load some webpacks, but is forbidden to do so. Inspecting the page shows a lot of js issues.
In Edit mode the number of Source-Map-Errors increases quite a lot.
Thanks Sven, escalating.
Best,
Jim Finch
Community Experience Manager
System Admin
@Jim Finch-Oracle
Old threads are still not locked
I thought this was a trivial thing to fix - and, indeed, I though it was fixed, as there have been no more in the last few days. But a five-year old thread in SQL & PL/SQL has just been resurrected this morning.
https://community.oracle.com/tech/developers/discussion/3798285/what-is-difference-between-is-and-as-in-plsql#latest
Cheers - mathguy
As did I, the fix was applied last week. Will follow up with the vendor.
Best,
Jim Finch
Community Experience Manager
System Admin
Sven W. wrote:
I'm extremely disappointed that for such a technical forum nobody seemed to have tested this important part before the forum went live.
At the original launch both inline and block code was displayed using a monospaced font. At that point the only problem I could see was random syntax highlighting of SQL and PL/SQL due to wacky misidentification as bash, Lua, Ruby etc.
Some time later (for thus far unexplained reasons) all font formatting failed for about 4 days. Part of the fix for that appears to include this CSS gem:
/* Source: https://community.oracle.com/tech/dist/forum/addons/oracle-plugin-theming.min.css line 506 */ * { padding: 0; margin: 0; box-sizing: border-box; font-family:Oracle Sans !important }This use of the universal selector and the
!importantdirective overrides all other font selection rules, resulting in every element on the site being rendered in the Oracle Sans face. This amateurish blunt instrument is preventing code from appearing as intended in various common monospace typefaces, as specified in this overridden rule:/* Source: https://community.oracle.com/tech/dist/forum/addons/oracle-plugin-theming.min.css line 5668 */ .userContent .codeBlock, .UserContent .codeBlock, .userContent code, .UserContent code, .userContent pre, .UserContent pre { vertical-align: middle; font-size: .85em; font-family: Menlo, Monaco, Consolas, Courier New, monospace; margin: 0; border:0 }Frankly I'm not surprised that Oracle have resorted to horrible hacks with this because I can't figure out what "Vanilla Communities" is trying to do in HTML and CSS either, other than it seems to be massively and needlessly overcomplicated (see also: performance).
Another interesting point is that despite the overwhelming push to present everything Oracle using the Redwood design language (including the APEX IDE UI 😡), they did not apparently commission an "Oracle Sans Code" monospace font to use in it.
@Jim Finch-Oracle
A 2009 thread, revived today.
https://community.oracle.com/tech/developers/discussion/1004480/convert-blob-to-varchar2-for-inserting-to-table#latest
More broken links:
Start from the developer community => Click on HOME (https://community.oracle.com/tech/developers/) => click on Connect with others
Result: https://community.oracle.com/tech/welcome/tech/developers/categories
To clarify. I think the problem is the link on the "Home" anchor has a trailing "/".
https://community.oracle.com/tech/developers/
instead of
https://community.oracle.com/tech/developers
which eventually leads to this behaviour. The two links should not behave differently, but they do.
Sent another e-mail to the vendor. They're working on it.
Best,
Jim Finch
Community Experience Manager
System Admin
Passing on to the designer. Thanks!
Best,
Jim Finch
Community Experience Manager
System Admin
Bundled up the thread and sent it to the designers and dev team.
Thanks everyone!
Best,
Jim Finch
Community Experience Manager
System Admin
The platform is getting completely unusable!
I am increasingly plagued with this:
Yesterday it took me about 30 minutes to write a message that should have taken less than a minute.
In a thread I opened in MOSC, one of the most respected, well known, published Oracle experts told me he was giving up due to getting the same.
I'm about to the point of asking people helping me on that thread to take the discussion to OraFAQ. At least they have a stable platform.
Even here, writing this message, I started getting it. And the "ask question' button was greyed (faded) as unavailable for a bit. Then, after finally posting, I decided to edit the post to add this comment, and the 'edit' button was equally faded, only beomeing available after repeated occurances of the 'web page is slowing' message.
I"m reminded of a line from 'A few Good Men', when Col. Jessup makes a disparaging remark about the type of unit being run there.
Just to document, when yesterday did you get the "web page is slowing..." message on MOSC?
I saw it on one of our testing sites this morning, but haven't seen it on a live site as of yet.
Best,
Jim Finch
Community Experience Manager
System Admin
@Jim Finch-Oracle
I got the "web page is slowing..." message several times over the past three days. Always when typing a "comment" - because, no doubt, the page script "does some things" in the background (such as saving what I typed so far). 100% on "a live site" (specifically, when typing comments in the SQL & PL/SQL forum).
I noticed that the save draft was being particularly aggressive during typing that last post, but not sure if that's the cause of it. I haven't seen the yellow on the live site since immediately after migration, where the categories were unsorted and unnested in the back end (500 unsorted areas is one thing I know to cause that). I'll try a few things with what I can do, then escalate on up.
Best,
Jim Finch
Community Experience Manager
System Admin
It's possible that the background saving is more time-consuming when we type something in a SQL editor (for testing), then we copy our code and the output and paste in our Comment. There is a lot of text added (pasted) all at once - which (depending on how the "save draft" is triggered, perhaps by character count or whatnot) may immediately cause the draft to be saved.
Just to document, when yesterday did you get the "web page is slowing..." message on MOSC?
I saw it on one of our testing sites this morning, but haven't seen it on a live site as of yet.
I'm getting it right now, 1:13 PM CDT, on another machine.
Thanks, I'll see if the devs are up to something.
Best,
Jim Finch
Community Experience Manager
System Admin
It might be. The fact that it's supposed to save every 30 seconds or so by default, and my last reply to Ed saved the draft in about 2 seconds (I'm not that slow of a typer), has me concerned. Of course, typing to you here saved normally. But I'm watching and it's also taking 2 seconds or so to bring the "post comment" back when it's saving.
Trying something now. Please let me know if anyone else is still getting the yellow on this node (the tech area). The saving seems to be less aggressive for me and the "post comment" is moving faster.
Best,
Jim Finch
Community Experience Manager
System Admin
Trying something now. Please let me know if anyone else is still getting the yellow on this node (the tech area). The saving seems to be less aggressive for me and the "post comment" is moving faster.
Ok, this post is just to test it out, FWIW because there sofar appears to be a random aspect. I am writing this at 6:45 pm US CDT. As I mentioned in a response in my thread on MOSC, I thought a commonality might be that I get it (the yellow warning about impacting my browser) only when I had pasted in a graphic or had formatted text as a quote or code.
One of the 'early symptoms' would be that keyboard input would be delayed, resulting in typing even full sentences with notohing being displayed, then several seconds later all of the input finally popping out. That has been happengin eveven as I type this sentnce and getting wrose as I go. Which ixplains the bad sepelling in this paragraphs. Also the 'post comment ' button woill be faded out as I thpe with to nofeedback.
BOOM, and theres' the yellow warning.
Another thing I notics is that at least on one occasion when I go tit the entire browser was impacted. Smply closing a tab took an inordinate ammount of thime. (that last sentence was keyed in entrirely 'blind'). even closing the entire browser was promblematic. Again that last sentence andthis wone is keyd in entireoly blind.
At this oint, I'm smlplu wating for the 'post comment' button to actuall be active so I cn post tis.
FWIW, my previous comment only got posted when at 7:08 US CDT, I closed my browser and reopend it, to find that it was still holding my comment, and finally the 'post comment button was active.
@Jim Finch-Oracle
I tried to reply to someone on SQL & PL/SQL and the site is freezing again for over a minute or each time the autosave kicks in or the browser is warning me of instability / unresponsiveness [Chrome 86.0.4240.75 (Official Build) (64-bit)]. It's pretty much making the site unusable for me, I can't type much before some huge delay or Chrome gives up thinking the site has become unresponsive.
It seems I'm not the only one who keeps seeing this issue. Methinks the site needs a better test cycle...
What is happening? Suddenly I went from Bronze Crown to Newbie. Now I got my old profile back, but clicking on a thread in my Notifications returns
Permission Problem
You don't have permission to do that.
yes, I'm sure there's problems - I was surprised to be able to open this thread. Though my profile page opens ok, with no untoward change.
All the pages I had open are now refreshing with that permission problem.
https://twitter.com/swesley_perth/status/1318699372549296128
One good thing: the blue dot for unread comments is gone after reading the comment.
I'd say that is a bad trade-off though! 🤡
Blue dot is back again, after reading your comment 😕
It's them negative waves, I say - woof woof 😅
At the moment I have only access to the community but no tech space
Hi,
I get the following message after login resp. when I've searched a subject/discussion
Any idea?
Regards,
martin
@Jim Finch-Oracle :
The groundbreakers is messed up, latest is from Jume?!
Somebody did develop in PROD... again 😂
Reported to vendor. It's both Apps & Tech and Developers.
Best,
Jim Finch
Community Experience Manager
System Admin
That concerns me, you should only be seeing the discussion list there. Not the polls. But you may be seeing different things due to your mod permissions.
Best,
Jim Finch
Community Experience Manager
System Admin
Confirming that there was an outage on the Vendor's side of things that resulted in the inablity for members to see posts on the Groundbreakers and Apps & Tech nodes this morning. This has been resolved.
Best,
Jim Finch
Community Experience Manager
System Admin
"This has been resolved"
Perhaps, but only partially. Why would I be able to access the site from my laptop (which I am), but not from my iPad?
This is something new. Last night's outage was simply not displaying posts. Users were able to login and otherwise use the site.
I'm able to sign in with my iPad. Are you attempting it over a VPN or anything with a go-between before accessing the site?
Best,
Jim Finch
Community Experience Manager
System Admin
Test account. Posting from iPad.
@Jim Finch-Oracle
Nope- straight over WiFi network at home. And it is fixed now, without any action on my part.
Re: last night, you may want to read the other people’s reports from yesterday in this thread. The problem was also related to permissions. See for example marwim’s post, a few above this one.
Mathguy,
Oh I've read the reports, and the twitter and the direct messages and... I've also been in the back end and talked to the vendor myself. There was an error on their end that caused a lot of wonkiness with a lot of things including member permissions. The roll back on their end seems to have cleared up the issue, but we're following up as to what they did and why it impacted the site the way it did.
Best,
Jim Finch
Community Experience Manager
System Admin
O...K...
Right now I am seeing the same SORRY, YOU’VE BEEN BLOCKED screen on my laptop, but not on my iPad, from which I am reporting this. Just a minute earlier I posted a reply from my laptop; I went to the top of the page and clicked on the category name, SQL & PL/SQL, to get back to the list of discussions, and I got the “you’ve been blocked” message instead.
Oracle screwed up monumentally with this migration. I am almost at the end of my patience; Stack Overflow’s gain!
It looks like it was a singular thread that was doing it. Thread has been removed and isolated to send to the vendor. I'm not sure if it was the code contained within the message that was setting it off but am escalating to the vendor.
Best,
Jim Finch
Community Experience Manager
System Admin
@Jim Finch-Oracle
I had the same access problems earlier today, totally locked out, then intermittently my account randomly turned into a newbie one with my whole history gone.
Is there any QA cycles going on before code is dropped in Production or do people just compile-it, ship-it? Because on the face of things it seems not much QA is happening, given the number of glaringly obvious and catastrophic bugs that anyone who spent about two minutes testing would see.
Some of this stuff is really development 101, e.g. people STILL can't format code in a monospace font. How many developers or websites do you know that use proportional fonts for the format option on code?!!! None, anywhere. It's fundamental on a site that has a huge amount of code / SQL / data samples posted, that require it. It's a simple change to CSS, and someone on this thread has even told your team where the CSS balls-up is.