Learned about Xfinity rewards providing a free year of access to
Perplexity.ai Pro, so I signed up. Ran a test by uploading a picture of a flower/weed in growing behind our condo and
produced this page.
#Based on the
reviews that I've read and seen, I think
the Pixel 9 Pro Fold is a very nice device. For me the key to any foldable is how heavy it feels when folded because I don't want to feel that my pants are falling down while carrying the phone in my pocket. The price is way too high for me, and I will never buy a phone priced above $700, however, as I am a heavy iPad Mini user who carries a smartphone, I believe that I am the type of person for whom a foldable is targeted. For the near term I expect to continue using the iPad Mini and smartphone because the combined cost is at my max but also because that cost is spread across two purchases at different times.
#
Good morning. We have now entered
meteorological fall, and therefore the change the header graphic.
#
I wonder how much
circle to search is used on Android. Google Lens makes sense, and I get the idea that circle to search might be useful, but I wonder whether it is truly useful. Maybe if one uses their phone to generally browse the web, but I mostly open web pages on my phone that I specifically sought out, and therefore know the information that I am seeking.
#Manton has added
AI generation of photo alt text and search keywords for all micro.blog users. Hopefully this will not drive up the cost of his hosting. Here is
an example of the generated result.
#
I wonder, does Best Buy have the Pixel Fold in stores? I would like to see the new Pixel Fold in person some time.
#I built a new remote desktop, again in an LXC container, running Fedora and XFCE. Speedometer 3 in Chrome nets a score of 10.7, which is close to the score I got when running Chrome on Windows 10.
#
Found out this morning that my new Proxmox host shutdown late Friday and I think it was due to the spinning disk getting too hot. The embarrassing part is I didn't know it was shut down until this morning when I tried to start my remote desktop, a reminder that I need to add this to my notification system.
#
I ran the Geekbench 6 benchmark on my Pixel 7a and it got a single core score of 1344 and a multi-core score of 3122, making it the second fastest single-core performance of my devices I've tested and third fastest on multi-core score.
#Ran the Geenbench AI 1 benchmark on my Pixel 7a with the TensorFlow Lite framework and CPU backend and it got a single precision score of 717, half precision score of 706, and quantized score of 1262. I then ran the TensorFlow Lite framework against the GPU backend and it got a single precision score of 532, half precision score of 616, and quantized score of 593. Finally, ran the TensorFlow Lite framework against the NNAPI backend and it got a single precision score of 273, half precision score of 2239, and a quantized score of 4991. I believe the NNAPI backend is the one optimized for AI, so the fact that it scored best is not suprising.
#
- Price for me is a big factor in deciding what smartphone to buy. I think $1,000 is too much to pay for a phone and $500 is my preferred high end. My preference is to pay less than $500 for a phone and I am willing to go over $500 by $200 if I am getting value in return. These prices are why I have been buying the A-series Pixel phones rather than the standard Pixel series. #
- Given price is a big factor, a correlating factor is trade-in value because that ultimately reduces the purchase price. Right now Google will give me $300 for trade-in of my Pixel 7a, which then brings the total cost of the Pixel 9 to $499 and the Pixel 8 to $399. What this means is if I wanted/needed the latest phone I could guy one within the price targets and I could buy last year's standard Pixel within my preferred pricing. #
- So then, then question becomes, is it worth replacing my one-year old Pixel 7a with either the Pixel 8 or the Pixel 9? The info I see says it's not worth upgrading to the Pixel 8 because all it provides is one generation newer processor and a slightly bigger battery, but worse camera specs. An upgrade to the Pixel 9 givens me much more: two generations newer processor, 4 more GB of RAM, a better ultra-wide camera, and finally support for the most recent Pixel features including AI. The better hardware is black and white, the value of the announced Pixel features and AI is gray at the moment. #
- Consequently, wait-and-see is the best strategy. Based on the trade-in value of the Pixel 6, I expect Google to provide that same $300 trade in value for another year and in that time there could be a price drop on the Pixel 9. Most likely Google will drop prices, or have sales, at Christmas time. I expect Google will stop selling the 7a and should drop the price of the Pixel 8 to at least $599, and Google could drop the price on the Pixel 9 that in combination with trade-in on the 7a results in a more appealing price.#
- What I want to see is whether the new Pixel features that end up not available for the 7a are really desirable. An approach is to wait until Google produces the Pixel 10 and drops the price of the Pixel 9, in other words, buying in on the N-1 Pixel. A factor is going to be whether Google continues the A-series. Rumors are they intend to drop the A-series and I fear the fact that they launched three Pixel 9 phones earlier in the year than normal lines up to that rumor. It would mean no new hardware at Google I/O and a summer release of new phones that are well in the supply chain for the upcoming holiday shopping season.#
Ars Technical included in
their review of the Pixel 9s a bar chart comparing Geekbench 6 scores of the Pixel 7a, up to the Pixel 9 Pro/XL. (They skipped the Pixel 8 for some reason). Interesting to me as a Pixel 7a owner to see the Pixel 7a overall multi-core score of 3,602 is not too much behind the Pixel 9 score of 3,885. The Pixel 7a single core score of 1,405 is actually better than the Pixel 9 single core score of 1,215. I wonder why the Tensor G4 is so poor at single core? What this tells me is that the Gemini AI stuff is more dependent on memory than CPU.
#
I wonder whether one can use a
Raspberry Pi 5 with a PiHAT & SSD connected to a
CrowView? It looks like that should be possible.
#After reading
Wired's article about Android 15, it looks to me like it mostly brings quality of life improvements and no real new features and this probably why Google didn't wait to include it on the new Pixels they just announced. I don't see many, if any, of the new Pixel features being dependent on Android 15.
#
- Because of the Google #TeamPixel kerfuffle this past week, I had great interest in reading the reviews of the Pixel 9 series posted by 9to5Google, The Verge, and Wired. The gist of the kerfuffle is, can one trust reviews when associated with a "#GiftFromGoogle" tag in light of leaked text of the latest agreement with influencers that sets an expectation that non-Google devices are not "preferred" over Google devices? Google's later clarification is, they don't consider #TeamPixel as a reviewers program, and I think implied in that statement is an expectation that participating influencers are influencing on behalf of the brand. I personally wonder why anyone thinks this is news, it should be obvious that brands, Google or any other, see influencers as extension of their marketing. What is an influencer anyway?#
- Back to the original question about trust, I think a related question is about value. When I read official reviews I can't help but feel the writers feel obligated to write something, anything, negative about the product being reviewed, which makes some sense because what product is perfect. Except, is what one writer think a negative really a negative? I think what you see in "tech reviews" invariably amount to biases. What happens is that products are reviewed against a writers preferences and not against some form of industry, or even publication, standard. #
- So, I think one needs to take tech reviews with a grain of salt. Tech reviewers of "recognized" publications are not necessarily any better or ethical than influencers. Many reviewers put high value in access to products and people of these companies. Like nearly every other piece of journalism, reviews are a mashup of facts and opinions otherwise the reviews would be nothing more than a print out of product specifications. At one time, publications like PC Magazine had labs to do performance testing and published the results against an established baseline and their reviews tended to provide explanation of what the deviations from the baseline mean. Of course opinions crept in to the explanations, they weren't nearly 100% opinions. Most tech reviews you read today are not a report of a product measured against a baseline, but rather a report of a person's experience about using the device and whether that experience met their expectations. (More like Chaos Manor and less like Byte Labs) The value of these reviews then depend on the degree to which the reader and the writer's expectations align.#
- I am a lifelong fan of the Chicago Cubs and feel blessed to have lived long enough to see them win the 2016 World Series. The players on that 2016 Cubs team will always be special and so when it came time for those players to part ways from the team I, like many, were disappointed, but I now have to admit they were the right decisions.#
- Last night Javier Baez, one of those on that 2016 who now plays for the Detroit Tigers, returned to Wrigley Field for the first time since he was traded to the New York Mets. Even when Baez played for the Cubs he was not considered a good hitter, he chased most pitches out of the strike zone, but some time had the good fortune of making a big hit. I always felt Baez's value was most defined by his fielding at shortstop, and in his prime I thought him the best shortstop in baseball. I am old enough to remember that the shortstop position was not usually fielded by the best hitters (remember Shawn Dunston?), but considered the most important in field position and Baez fit that mold. #
- Even though I did not like the fact that the Cubs seemingly unceremoniously dumped those players from the 2016 roster, I have to admit that looking back now nearly all were the right decisions, except for one, the first one letting Kyle Schwarber go as he went on to win another Word Series with the Nationals and play in another with the Phillies. #
- Kris Bryant has not played a full season since he was traded to San Francisco. Anthony Rizzo has been alright but often injured with the Yankeess. Baez has never figured out the plate. It's clear to me now that even if they had kept these players the Cubs would not be any better of a team than they are today. #
- So, the front office was right in letting these players go for as many prospects they could get. Some, like Pete Crow-Armstrong, who they got in the trade to the Mets for Baez, has potential to be stars in their own right. My fear though is that the current Cubs front office has a profile for players that seem to be one hit wonders over long term producers. In short, the new guys they get are like the old guys they had and that seems to produce the same results of not being good, nor clutch, hitters. You combine this hitting profile with a tendency turn over the bullpen every single off season and you get the mediocrity that is the current adjective for the Chicago Cubs. #
I did not know that weather radar decreases detecting events like tornados and lake effect snow, which typically are below 4,500 feet, at about 60 miles. The beam of the radar leaves at a half degree angle and that increases the further it gets away from the radar so that at 120 miles the beam is at 12,000 feet up in the atmosphere. To increase the effectiveness there needs to be
more weather radars deployed. The National Weather Service has four radars in Michigan (Pontiac, Grand Rapids, Gaylord, and Marquette), which if you use Radar Scope they are prefixed by the letter K. Radar Scope also shows the smaller radar from Detroit Metro Airport (prefixed by the letter T).
#The Go version of Fabric has been released and I performed the migration on the debian12.home container. I need to do the same on my Macbook.
#
Napkin is an app that creates visuals from text that you input. You don't write a prompt, like with other generative AI tools, you paste in text that you want turned in to a visual and the app provides you different results to pick from.
#
Google might be canary testing a new AI feature called Theft Detection Lock (
via Forbes):
#
So, what is theft detection lock? The Google announcement from Frey refers to it as being an “automatic AI-powered screen lock for when your phone is snatched,” which sums it up pretty well. Essentially, theft detection lock uses Google’s AI to determine if someone has snatched your smartphone by identifying the kinds of motion most often associated with such theft: the snatch from hand followed by escape on foot, bike or car. If it considers the motion to be indicative of a theft, the Ai-powered feature will instantly lock your device to prevent the thief from easily accessing your data.
#I wonder how many false positives can be generated by trying to detect the type of motion associated with theft. If is sudden acceleration, what happens when I am driving my car and pull out to pass?
#
- Ran Geekbench 6 in more containers today to compare performance. First I ran it in the Fedora 39 LXC Container in Proxmox 8.2 on the Raspberry Pi 5 and compared that to the Debian 12 LXC Proxmox container running on the Beelink U59 Pro and the results support the better performance that I have been observing. This comparison is a bit apple to oranges given the different Linux distros, but I think it generally supports the notion that the ARMv8 processor used by the Pi is slightly better than the Intel N5105 even though the Intel processor has a slightly higher clock speed. The performance of the Fedora 39 container is nearly the same as as the Raspberry Pi OS, and in fact the single core score is slightly better. #
- The Debian 12 LXC Container, which has been my main remote desktop before I built the Raspberry Pi 5 with the NVMe drive, performs significantly better running in Proxmox 8.2 on the HP EliteDesk 800 G3 than in Proxmox 7.4 on the Beelink U59 Pro, and this performance improvement is expected given Intel i5-6500 3.6 GHz processor that is in the EliteDesk. From a simple spec and performance perspective the EliteDesk is more than twice as fast in single core performance and a 5 times faster in multi score performance and I paid less for the EliteDesk, even though it is physically much larger. BTW, the $106 I paid for the EliteDesk is also much cheaper than what I paid for the Raspberry Pi 5, which is all to say that if one is building a home lab on a budget then the refurbished route is probably the best one to take.#
The second attempt at installing Proxmox was successful, and I know have a Proxmox Version 8.2.4 host in the house!
#Installation of Proxmox on to the 1 TB NVMe is taking a long time. I've found others who have reported the same issue, sits at 3% on LV install for a long time before quickly finishing up. The first install completed and rebooted the computer back in to Windows, and I saw problems with Secure Boot that seemed to prevent the installation from working. I am installing a second time after turning Secure Boot off.
#I received the HP EliteDesk 800 G3 that I ordered via Newegg and was surpised that it was the larger "Small Form Factor" model rather than the smaller mini computer model. On one hand, the larger model is easier upgrading, but on the other hand it will take up more physical space. Last night I booted it up, configured the Win 10 that it comes with, download and ran Geekbench 6 and found it to be the second fastest computer in the lab.
#The Geekbench 6 scores of the EliteDesk are twice that of the
Beelink U59 Pro that is my "new" Proxmox computer, and it only cost me $100 (refurbished and very clean). My original plan was to use this EliteDesk as a backup server, but it seems foolish to not use this better performing computer as a Proxmox host, so I am installing Proxmox 8.2 on it and plan to re-purpose the HP Probook 6460b that is my first Proxmox server as the backup server.
#
Geekbench has released an AI benchmark, which I suspect is going to become a benchmark baseline for personal computing. Don't know what the scores mean, I have to read about them, but my
Macbook Air got these: single precision score 2836, half precision score 4294, quantized score 3533. The framework was Core ML against the CPU backend. For some reason the results are not stored on my profile so I captured the results in Archivebox.
#
Gemini Advanced costs $20 per month, the same cost per month as the
chatGPT subscription. The $20 price tag feels to me to be a barrier to entry of this level of AI for most people, who already may a monthly price for their mobile carrier.
#I wonder whether Google Gemini is now able to work with Google Home routines. The fact that I cannot tell Gemini to trigger a routine is why I have not installed Gemini on my Pixel 7a. Update:
Tom's Guide says it can, so I am going to try installing it. Despite what Tom's Guide says, I cannot ask Gemini to trigger a home routine. I am going to keep it on for now to see if an update is pushed.
#9to5Google speculates that the new Pixel Weather app will come to "old" Pixel phones, like my Pixel 7a, in the December Pixel feature drop.
#
I wonder when Google will release a new
Pixel Tablet with the processor and specs to handle their AI?
#
- Thinking about the debate between AI and photography. I've heard the question, "what is a photo?" being raised, implying a definition of a photo as a graphical capture of a real moment. The introduction of AI tools that allow one to alter photos from something that really "happened" to something that didn't is what is being considered a problem. Except, what about how Photoshop has been used all these years? Presumably, people see editing a photo to improve it is different, but that editing can turn into a photo of something that did not happen. What about Google's Magic Eraser? #
- Perhaps there ought to be a definition of what is produced when one uses AI to alter a photo. I would suggest calling them composed pictures. In my mind a photo that is "re-imagined" into something else is more like a painting than a photo and I think I would refer to them as something other than a photo. I think we are seeing the development of a new art form, the ability to craft the right words to generate a picture.#
- While I personally would never present a "re-imagined" picture as a photo, I think there needs to be at least a best practice where these AI tools include some type of watermark indicating they were produced by AI. The whole idea of discerning that which is real (news, photos, videos) and that which is fake is a problem because humans are lazy. I personally want to trust people and think living in a world in which everything has to be verified is exhausting.#
Google is offering $300 trade-in on my Pixel 7a, which would bring the price of the Pixel 9 down to $499. which is the same price as the Pixel 8A. I do hope that Google keeps selling a phone at the $500 price target.
#Surprised that Android 15 was not released during today's Pixel event, in fact
the Pixel 9 series ships with Android 14 installed, which is a really odd decision; normally the new phones ship with the new version of Android. All reports indicate beta testing has gone well with Android 15, so there has been good reason to expect it ready by now, but with it not ready I wonder why then they moved up the Pixel event? Might this be a precursor to Google dropping the A series as expected? Moving up the Pixel event to earlier in the year would compensate for not announcing a new phone during Google IO.
#Long elections are all about money, pure and simple. Hopefully one of the lessons to be learned from this upcoming U.S. Presidential election is that they don't have to be so long. We really need election reform in the U.S. that is intended to make them more efficient, I personally advocate capped public spending. We don't need billions spent on commercials. I also would like Congress to pass a law that effectively overturns the SCOTUS Citizens United case by declaring spending money is not free speech.
#
Last week I started using a LXC container, hosted in Proxmox on a
Raspberry Pi 5, running Fedora 39, XFCE as my remote desktop. Using Chromium, it gets the highest Speedometer 3 score of my computers other than the
Macbook Air. After using it for a week, I happy with and surprised by the performance. On Friday I briefly tried my old Debian 12 container hosted on the
Beelink U59 Pro and while Chromium runs well, it still felt a bit sluggish, although that is not a true Apple/Apples comparison because not only us that Debian versus Fedora but also KDE versus XFCE.
#
And then, once everyone cleans up and shakes the debris off their phones and laptops, so much of what Trump said seems too bonkers to have come from a former president and the nominee of a major party that journalists are left trying to piece together a story as if Trump were a normal person. This is what The Atlantic's editor in chief, Jeffrey Goldberg, has described as the “bias toward coherence,” and it leads to careful circumlocutions instead of stunned headlines.
#The bias toward coherence... like that.
#I have a hard time believing there are really "undecided" voters at this stage, at least in deciding between the two leading candidates. People know Trump and are either for him or against him and had the same opinion in 2020. Maybe there are people deciding whether to vote for either or some other third party candidate, or not vote at all. I honestly think that too much attention is spent on the Presidential race and not enough on the races for Congress, particularly the Senate. Not much gets done without the Senate.
#
The only censorship that Musk is really concerned about is censorship of him, which is why he bought Twitter. The purchase of Twitter by Musk has always been about providing him with a bullhorn, and the only response one should have to this is to stop using Twitter/X. No matter if X is the place where breaking news is first found, the only real responsible response to it is to stop using it.
#
A fair question about Google being a monopoly, what is the consumer harm? Is the ruling changing the purpose of antitrust law in the U.S.? Perhaps the question about harm comes more in to play during the remedy phase?
#The antitrust ruling against Google describes how Google benefits from being a monopoly, but
Om Malik convincingly describes how Google's monopoly is it's undoing. Thinking back to the Microsoft antitrust case in the 90s, I can't help but wonder whether going through that helped Microsoft in the long run as it seems to be in a better position today than expected.
An article in The Atlantic points out that how the U.S. handles antitrust is too reactive and therefore doesn't prevent harm that is described in the ruling, instead there ought to be greater scrutiny and blockage of mergers and acquisitions.
#
I learned about the
Obsidian Second Brain plugin and decided to try it out with Obsidian running on my
Macbook Air. I have it configured to use Ollama and a the phi3 LLM but so far it hasn't successfully completed indexing my vault, which contains all of my exports from Readwise. I have a feeling the
Macbook Air doesn't have enough horse power.
#
I have been experiencing frequent disconnects using RDP to one of my remote desktops using the Windows 11 Remote Desktop Client and today I found
a potential fix/workaround.
#I had a productive weekend tech wise, built a new remote desktop using an LXC container running in Proxmox on the
Raspberry Pi 5 and I installed PagePark on a new server to host my outlines.
#
- I work from home and do not want to do any personal stuff on my work computer. While I can use my Macbook, I find it more convenient to use a remote desktop client on my work PC and have it open as a window on my work computer. Nearly all of my personal computing is done using a web browser, so browser performance is most important to me and I am using the Speedometer 3 benchmark to measure that performance.#
- The Macbook blows away performance of any other computer I have access to, but for the most part I don't really need that level of performance to browse my RSS feeds or read articles in Readwise Reader. What I am using are Linux desktops, mostly KDE Plasma, XFCE, or LXQt, running in a container or virtual machine running in Proxmox on either a Beelink U59 Pro or a Raspberry Pi 5 that is on my home network. #
- The baseline performance I seek is better than or equal to the Speedometer 3 score using Chrome on my work PC, which is an abysmal 4.04. For comparison, the same benchmark score using Chrome on my Macbook is 22! Yes, my work PC sucks. #
- For the last six months I have been using KDE in a Debian 12 Proxmox LXC container on the Beelink U59 Pro on which Chrome scores 2.63, not great but better than trying the same configuration in a virtual machine. A while back I bought a Raspberry Pi 5 which I recently "upgraded" by adding the Pi NVMe HAT and a 1 TB NVMe SSD. My first testing involved installing the Raspberry Pi OS and Chromium that achieved a Speedometer 3 score of 4.53, which is better than the work PC. #
- I could have just left the Raspberry Pi OS on that board and connected to it with VNC or RDP, but I wanted to try out Proxmox on the Pi 5 to see how it would perform, which I have done. Given that the Raspberry Pi 5 uses an ARM processor, that limits what one can run on it to virtual machines or containers supporting the ARM (aarch64) architecture. So far I have been able to run Debian 12 and Ubuntu 22.04 virtual machines, and I have tried LXC templates for Fedora 38, Alpine, Rocky Linux, and Ubuntu. #
- Over the weekend I provisioned a Fedora 38 container with 4 vCPUs and 4 GB of RAM, upgraded it to Fedora 39, then installed XFCE and Tigervnc. Again, being this container runs on ARM, the browser options are Firefox and Chromium as I haven't found an ARM version of Chrome that I can download and install in Linux. I was surprised to see a Speedometer 3 score of 4.04, that is much better than performance of the Chrome + Debian 12 container I had been using. #
- The Debian container has 2 vCPUs whereas the new Fedora 39 container has 4 vCPus, so I wondered whether adding 2 more vCPUs would make a difference, and it did, increasing the score to 3.73 from the 2.63 score I had seen previously. I next installed Chromium on that Debian desktop, ran Speedometer 3 again and it got a score of 4.73! #
- The results align to expectations in that the performance of the Intel Celeron N5105 CPU and the ARM Cortex-A76 have comparable clock speeds and 4 cores. Chromium appears to be that much better than Chrome, and that is noticeable in these more constrained performance environments. #
- I do have one apples-to-oranges situation in that I am using different distros, desktops and connection software between the two containers. I doubt that desktops (KDE and XFCE) or connection (RDP and VNC) software will make that much a difference in browser performance, but I wonder about the distro. Unfortunately, I can't seem to get any Debian ARM LXC template provided by Proxmox to work on the Raspberry Pi, so that means I will have to provision a Fedora container on the Beelink, which I will probably do and install XFCE and Tigervnc on it to see what results I get.#
Daniel Miessler suggests replacing the phrase "Artificial Intelligence" with "Intelligence Tasks." The explanation he provides for what are Intelligence Tasks does a good job of describing how this new technology is more than automation. I recommend reading and saving this article.
#
An interesting quirk with Proxmox on the
Raspberry Pi is that it doesn't show the memory used by a container like appears on Intel hosts. If I assign 4 GB of RAM to a container, I see how much of that 4 GB of RAM is being used by that container with Intel, but on ARM (Raspbery PI) Proxmox doesn't show any memory used and when I run htop within the container it shows the total host RAM and not the RAM assigned to the container.
#Disabled smooth scrolling in Chromium and that seems to make things a bit snappier.
#Welcome to the first day of August, 2024! Today I am using a new virtual desktop running in a Fedora 38 LXC container that is hosted on a
Raspberry Pi 5 running Proxmox. Speedometer 3 browser benchmarks are a little faster in Chromium with in this container than in the Debian 12 container I have been using that runs in Proxmox hosted on a
Beelink U59 Pro.
#
I have built another remote virtual desktop on my home network using a Fedora 38 LXC container running under Proxmox on a
Raspberry Pi 5. Right now I am in the process of setting up Chromium for web access.
#
- Officially the All-Star game is considered to be the half-way mark of a MLB season, but the real mid-season milestone that matters is the trade deadline. The actions taken by a team by the trade deadline indicate whether the team management thinks it can make the playoffs, if they do then they will try to make trades that help the team win in the current season, otherwise they start work on the next season. Jed Hoyer's comments that he was focusing on 2025 was indication that Cub's management doesn't expect to make the playoffs this year.#
- Last season the Chicago Cubs were surprising buyers at the deadline, having gone on a winning streak that convinced Jed Hoyer the team had the shot at a wildcard spot, but that didn't happen and David Ross the scapegoat. Now this season, after signing Craig Counsell to the highest salary of any MLB manager the Cubs are last in the NL Central with a 51-58 record and they are seven games out of the last wild card spot. In short, the Cubs are likely not making the playoffs this year. Are they going to fire Counsell?#
- Even though the playoff positioning is different between last year and this year, a fair assessment is that the team is really no better than the year before, nor the year before that or any other year up to 2016. What we have is the same problems of unreliable hitting and pitching from the bullpen. The fact that the 2024 team's starting pitching has been so good is as much an indicator of how much MLB has changed since 2016 than anything. It used to be that teams with strong starting pitching were winners, but the game now does not rely so much on starting pitching to win games. #
- The sole purpose of starting pitching in the game today is to not lose the game in the first five innings by not giving up more than three runs, the bullpen is expected to now win every game by not giving up any runs. If that weren't bad enough, hitting is a completely lost art in baseball, with hitters needing only a .244 batting average to be considered good. The game has come down to this, pitching keeping the opponents from hitting home runs until your own batters hit two or more home runs. #
- The Chicago Cubs line up has been made up of the similar type of unreliable hitters for nearly a decade now, which is a clear indicator to me that the front office's idea of what is a major league hitter is out of sync with reality. As long as Hoyer is running the Cubs they will not be any different, and so it's on ownership to recognize the insanity of continuing to do the same thing and expecting a different result. #
- If I were the owner of the Chicago Cubs, now that the trade deadline has past, I would fire Hoyer immediately to signal that it's time for a change. Take the time to review and interview people over the remainder of the season to find the right replacement to have your new front office in place before the off season begins in earnest.#
In her Substack today,
Nadia Bolz-Weber writes about how we self-identify and that who we are is different from what we like. I know there is a difference between what we like and who we are, and those who read my writing about religion may say, "I like the Green Bay Packers" but I wouldn't say that they are a part of my identity. For some this is true, but I think it comes down to how much loyalty you give that that particular thing.
#