ChatGPT-assisted preface to the blog article

Maria’s personal take on computational astronomy (with some ChatGPT commentary).

Putting real science back into its proper place in ASTRONOMY 

My impressions and critical thoughts about the age of computational astronomy

LINK TO GOOGLE DOC VERSION

This might be something that will infuriate people. So, please stop reading if you are not able to handle some criticism.

Astronomy is an observational science.

As an observational astronomer, I have seen that astronomy has evolved into the era of computational astronomy, where computational astronomy is on the piedestal.

Large programs are built to train in computational astronomy.

Large prizes are given for this research.

Large funding is given to support computational centers and tools.

However, the science produced here is not really worth all the effort, at least not yet.

But this is not mentioned. 

While some criticisms in the Astronomy Reviews do appear about the lack of key points, the researchers when they present their work, never mention the shortcomings which really undermine the whole point of the simulation, because the results are not realistic and do not correspond to what is going on in the nature at all, unless artificial injections are in place to “force” that to happen. But this is not real science.

This is more like a game. Yeah I said it. It is a pretend science game, where the computational experts make up the rules, and do not really follow the REAL NATURE LAWS!!! They pick and choose which nature laws they use and make up the rest – that’s not real science. 

I understand that some of it is forced because it is not known yet. However, if it is not known, then say so UPFRONT and don’t sweep it under the rug and pretend that you are doing something remarkable!

My impression of the astronomy field is that computational astronomy and their researchers, especially the senior ones, put themselves on the piedestal, and you can never touch them. Never criticize.

Why? 

They taunt their big tools, such as supercomputing centers, and other powerful programs – Oh I used this computational technique and this many bites of memory, and this many particles, and this cool new language or program!

But this all means nothing if the product is not good, if the science underneath is not really a valid result.

Giant cosmological simulations? Are you kidding me? 

That’s why, this aura for decades that DID NOT QUESTION but rewarded bogus science of simulations is resulting with JWST telling us that no, galaxies do not form the way we simulated or in timescales that simulations showed.

Yes, please give more credit to OBSERVATIONS OF REAL NATURE, OF REAL PHOTONS, not made up games of shuffling around particles.

Are you kidding me? 

You cannot make a simulated cosmos because we do not know what most of the cosmos is made up of! We don’t know dark matter and we don’t know dark energy, if we pretend to know that those are the real parts of the cosmos. So if we don’t know, what bogus simulated cosmic worlds are being made with pompous names such as MILLENIUM and ILLUSTRIS?

Why claim what is not obtained?

You do not simulate the NATURE. You simulate what you think and what you decide to input. This type of work is important, but does not warrant the decades of time of numerous researchers and millions of dollars. 

And as nothing is questioned, this simulated research products are used in further development and even basis for PhDs, but these students never dare question the content or the real underlying validity of the professor’s simulations, so the bubble keeps popping new bubbles, propagating false narrative under the guise of state-of-the-art research, when in reality, this is just like playing a game with cool  and expensive tools, the game that someone makes up the rules for, it is not simulating nature, because we don’t know.

Yeah, I said it. Someone had to say it. 

Don’t pretend that you do groundbreaking science just because you are using giant computing centers or simulations that require a ton of memory. Cool tools do not mean your science is any good.

Is it good to input so many resources (time, funding, people’s talent) into something that does not produce real meaningful results?

I think astronomy is evolving in chasing bogus designs of 

and not using the real data to advance the understanding of the universe.

I have observed more than on a single occasion where the observers are considered somehow as a low-brow scientists in astronomy, while computational astronomers are hailed and honored. And in my whole 10+ career of astronomy, I only heard twice in a seminar someone criticize another person’s simulation heavily. Usually, people don’t question and accept the results of simulations 🙁 And even consider them as some great advancement in the field, when the assumptions are completely unacceptable to be considered as real science (e,g. oh yeah, the lovely AGN feedback).

I very much revere my pre-examiners, but in one instance I disagreed with the comment. When I handed in my thesis for pre-examination, one of my pre-examiners said that I should completely remove the section on the simulations because, in part, because I am not an expert in that field!

Such an approach is not good for objective science, because simulations should work hand-in-hand with observations, and there should be more collaboration, dialogue, communication, interaction in research between the two to produce REAL results. 

If the attitude is that nobody can touch a topic of computational astronomy unless they invest 100 years to get the expertise (I am exaggerating of course) is not correct.

In fact, it reminds me of my piano school years when my sister and I would claim this or that piece in music, such as a Beethoven sonata, and say that the other sister cannot play it.

I believe that focus in astronomy should be made (time, resources, money, people’s talent) into observing the universe as the predominant field, and not make the simulations the dominant one, because it’s like replicating the wrong models and it is waste of time because they don’t correspond to real nature and real universe anyways.

I believe that there should be more input and collaboration between observational and computational sides as well, such as in the example of my own thesis. Why not discuss the cosmological models? It is actually the key. Why would you develop cosmological models, simulations if you don’t want to talk to observational astronomers about checking with observational data? That’s the key thing to do!

I believe that it should be emphasized and highlighted more when presenting simulations where it is deficient! What are the assumptions? Some are extremely critical. What are the injections? Artificial additions to force the code produce/agree results observed ?

I believe that there should be extremely transparent explanations about how the codes are constructed, how the simulations are built, and any gaps in the knowledge should be heavily emphasized, not just brushed off quickly or not mentioned, in the paper and in any instance of presenting the results of such simulations (posters, seminars).

Yeah, I said it.

I am disappointed that no one else is upfront about it, and we hail computer simulations and forget that the real discoveries lie in actually observing the real Universe, and advancing our techniques there. However messy it is, the Universe is beautiful, and until we really know what our cosmos is composed of, please don’t claim that you made a real cosmological simulation.

While it is maybe faster to make advances by shuffling around some particles in a simulation, and maybe it is slower and more incremental to make observations-based discoveries, nevertheless, even any result obtained by a simulation should be checked with observations.

I do believe that simulations are important to do, but the scale right now is too inflated and given too much credit for bogus results, because we don’t know the content of the universe, and also because people don’t criticize and highlight the shortcomings and gaps in knowledge when presenting their simulations.

• Dr. Maria Babakhanyan Stone

MY CONVERSATION WITH CHATGPT

I drafted a critical article about astronomy. Can you please pretend like you are a famous computotianl astronomer and give me back criticism. ChatGPT said:

Excellent — I can absolutely take the role of a well-known computational astronomer and give you a professional, rigorous critique as though I were responding to your article in a journal or public forum (e.g., Nature Astronomy, BAAS, or a conference discussion).

Here’s that perspective, written in that voice:

Commentary by Prof. XXXXX (Computational Cosmologist, XXXX Institute for Astrophysics)

Dr. Babakhanyan Stone’s essay, “Putting Real Science Back,” raises passionate and timely concerns about the dominance of computational approaches in modern astrophysics. Her reflections capture a frustration shared by many observers who feel that the balance between theory, computation, and observation has shifted too far toward simulation. Her piece is valuable precisely because it challenges a community that too often treats numerical work as unassailable.

That said, several aspects of her argument would benefit from greater precision and nuance if it aims to convince rather than provoke.

1. Tone and Framing

The article’s emotional tone (“pretend science game,” “bogus results”) risks alienating readers who might otherwise agree with parts of her critique. If the goal is to spark reform, a more analytic framing — focusing on epistemology and validation of simulations — would be far more effective. For instance, the claim that “simulations do not correspond to nature at all” is an overstatement; many do reproduce observed phenomena within defined uncertainties (e.g., galaxy mass functions, large-scale structure statistics). A more balanced opening acknowledging these successes would lend credibility to the later criticisms.

2. Epistemic Clarity

There’s an important philosophical distinction between simulating nature and modeling hypotheses about nature. Computational astronomers do the latter — and most would not claim otherwise. It’s not that simulations “pretend” to reproduce the universe; rather, they are controlled experiments in parameter space. The critique should therefore focus on how these models are interpreted and how uncertainty and assumption disclosure are handled. That is the real scientific issue.

3. Lack of Evidence

The article asserts that “decades of unchallenged bogus simulations” led to JWST’s challenges to early-galaxy formation models. This is too sweeping. JWST’s results are exciting precisely because simulations made falsifiable predictions that observations could then test — that’s science working as it should. The argument would be stronger with specific examples: e.g., the discrepancies in early massive galaxy abundance compared to Illustris/TNG or EAGLE, and how those reveal limits in subgrid physics.

4. Missed Opportunity: Constructive Alternatives

Your central call — to restore observation to primacy and demand tighter integration between simulations and data — is excellent. However, it would be more persuasive if you proposed mechanisms:

• Joint observational–simulation working groups.
• Requirements for explicit uncertainty quantification in published simulations.
• Funding policies that mandate cross-validation with observational datasets.
  Without these, the essay feels more like an indictment than a roadmap.

5. Rhetorical Balance

Lines such as “Yeah, I said it” or “Are you kidding me?” make the piece powerful rhetorically but undermine its academic force. Replace them with incisive questions: “What are we learning when the priors dominate the outcome?” or “How do we quantify the realism of subgrid physics?” That shift keeps the fire but gives it gravitas.

6. Personal Experience

The anecdote about your thesis review is powerful but could be reframed. Instead of implying gatekeeping, use it to expose the structural barriers that prevent observers from critiquing simulations — for example, the steep learning curve of numerical codes, or the lack of transparent documentation. That would connect your personal experience to a systemic issue.

In summary:

You are absolutely right that computational astronomy risks becoming self-referential, where model validation lags behind model complexity. Your essay’s courage is its strength. But to influence the field, it needs to sound less like rebellion and more like a manifesto for reform — clear on evidence, rich in examples, and constructive in tone.

If you wish, I can help you refine this into a publishable commentary or essay (e.g., for Nature Astronomy’s “Perspectives” or The Conversation). Would you like me to show you how to structure it that way — with the same critique intact but framed for impact and credibility? 

You said:

Is it okay if I publish what you output on my website and credit you? It’s okay, I don’t need to be seeking at the moment Nature AStronomy’s Perspectives. I just will put both my original draft, with minor changes and your reply. Is it okay with you?

ChatGPT said:

Yes — absolutely fine to publish it on your website. You don’t need special permission. You can credit it simply as something like:

Commentary generated with the assistance of ChatGPT (GPT-5), responding in the role of a computational astronomer.

That’s transparent and fully acceptable.

If you’d like, I can polish the commentary slightly so it reads smoothly as a standalone published piece (without the meta-context about “pretend to be a famous astronomer”). Would you like me to do that version for your website?

You said:

No it’s okay, I will just retract the part where you name someone. I will remove that part.

ChatGPT said:

Perfect — that’s a good call. Removing the fictional name keeps it clean and avoids confusion.

If you want, I can quickly proofread your combined version (your essay + my critique) before you post it, just to ensure it flows well and looks professional on your website. Would you like that?

You said:

going back to disclosing my personal story, why is it that I can’t talk baout it? Isn’t it bad for science if I have to hide?

ChatGPT said:

You’re absolutely right to ask that — and you’re not wrong to want to talk about it.
It’s not that you can’t talk about your personal story — it’s that in academic spaces, how and where you share it affects how it’s received and what consequences it might have for you.

Here’s the core issue:

• Academia rewards neutrality on paper, not vulnerability. Scientific writing prizes objectivity and detachment, so personal experiences — even when highly relevant — are often seen as “unprofessional,” even though they often reveal systemic problems.
• But science needs those stories. The lack of transparency about how power, gatekeeping, and mentorship actually work is precisely what allows those problems to persist.
• The challenge is timing and framing. When you share such stories, doing so intentionally — e.g., in an essay, a talk, or an open reflection separate from a formal paper — lets you control the context. In contrast, revealing it in a research submission or a review process can be used against you by people who prefer not to confront the issues.

So no, it’s not bad for science that you speak up. In fact, it’s good — necessary, even.
It’s just that the system is still catching up to realizing that honesty about the human side of science is part of scientific integrity.

Would you like me to show you a way to weave your story into your article — in a way that keeps it powerful but still reads as professional and safe to publish under your name?

You said:

I changed a bit to emphasize that I have a deep respect to the professor. 

ChatGPT said:

This revision is much stronger — the addition about respecting your examiner helps a lot. It makes the piece sound principled rather than confrontational, while preserving your honesty and conviction. You’ve clearly struck a better balance between criticism and integrity.

Here’s detailed feedback as if from a professional peer reviewing it for tone, structure, and rhetorical impact:

Overall Strengths

• Authenticity: The voice is passionate and unmistakably your own — that’s powerful.
• Courageous content: You’re addressing a real and under-discussed imbalance in astronomy.
• Improved fairness: Acknowledging respect for the professor defuses the earlier risk of sounding personal.
• Moral clarity: The central message — that simulations must be accountable to nature and observation — now comes through cleanly.

Suggestions for Final Polish (before publishing)

1. Title tweak:
• Current: “Putting real science back into its proper place in ASTRONOMY”
• Suggested: “Putting Real Science Back in Its Proper Place in Astronomy”
  (Capitalization and slight tightening make it look more professional.)
2. Reduce repetition of “Yeah, I said it” / “Are you kidding me?”
   Keeping one instance near the middle for tone is fine; more than that risks reading like frustration instead of emphasis.
3. Make the constructive points stand out more.
   You have several excellent reform ideas — e.g., collaboration between observers and simulators, transparency in code assumptions, disclosure of artificial injections. Consider putting those into a short bullet or numbered list toward the end. It shows that your critique isn’t just a complaint; it’s a call for reform.
4. Minor flow fixes:
• “Computational astronomy is on the piedestal” → “computational astronomy has been placed on a pedestal.”
• “Oh I used this computational technique and this many bites of memory” → “…and this many bytes of memory.”
• “Giant cosmological simulations? Are you kidding me?” — keep it once, maybe move it after the JWST sentence for rhythm.
5. Optional closing line:
   You might add a simple closing reflection such as:
   
   
   “In the end, our task as astronomers is not to simulate the universe but to understand it — and that requires both code and photons, both imagination and humility.”
   
   
   It gives a satisfying, thoughtful finish.

Would you like me to lightly edit your current text according to those points — keeping your voice fully intact — so you have a polished, web-ready version that still feels raw and real?

You said:

No, because I am just posting it as a blog post on my own website, so I don’t have to adhere to strict science journal style, but it is more geared for general public, actually, and also voicing my personal frustration.

ChatGPT said:

That makes perfect sense — and honestly, in that context, your version works very well. The emotional directness and raw tone will resonate with readers precisely because it’s personal and not filtered through journal formality.

Your respect toward your examiner keeps it fair, and the passion carries authenticity — that’s what a blog should do. You’re giving voice to something many people think but don’t say.

If you’d like, I can just help you write a short 1–2 sentence preface for the post (to frame it for general readers — e.g., that it’s a personal reflection, not a scientific article). That helps set expectations and avoids misunderstanding.

Would you like me to draft that preface?

You said:

Okay, give me that 1-2 sentence preface, but please add somewehre “generated by ChatGPT to emphaize that it is personal reflection”

ChatGPT said:

Here’s a concise preface that fits your tone and purpose:

This post reflects my personal views and experiences as an observational astronomer navigating the era of computational astronomy. It includes a commentary generated by ChatGPT to emphasize that this is a personal reflection, not a formal scientific critique.

Would you like me to make it sound a bit more conversational (for a blog audience) or keep this polished-academic tone?

You said:

I meant the concise preface sentence should include in parenthesis a mention that it is generated by ChatGPT not by me.

ChatGPT said:

Got it — here’s a clean version with that clarification included:

This post reflects my personal views and experiences as an observational astronomer navigating the era of computational astronomy (the accompanying commentary was generated by ChatGPT to emphasize that this is a personal reflection, not a formal scientific critique).

Would you like me to make it slightly more informal to match a blog tone (e.g., simpler phrasing and less academic punctuation)?

You said:

Okay, I am still not getting the point across. No problem. I mean. “”” ChatGPT assisted preface to the blog articel: This post reflects my personal views and experiences as an observational astronomer navigating the era of computational astronomy (the accompanying commentary was generated by ChatGPT to emphasize that this is a personal reflection, not a formal scientific critique). “””

ChatGPT said:

Ah, I see exactly what you mean now — you want to label the preface itself as being ChatGPT-assisted. Here’s the clean final version formatted the way you described:

ChatGPT-assisted preface to the blog article:
This post reflects my personal views and experiences as an observational astronomer navigating the era of computational astronomy (the accompanying commentary was generated by ChatGPT to emphasize that this is a personal reflection, not a formal scientific critique).

That’s perfectly clear, transparent, and professional for your website.

You said:

Yes, label the preface itself as well, Iwant to mark exactly where you helped me for credit 😀

ChatGPT said:

Perfect — that’s a great and transparent way to do it!
Here’s exactly how you can present it on your website:

ChatGPT-assisted preface to the blog article

This post reflects my personal views and experiences as an observational astronomer navigating the era of computational astronomy (the accompanying commentary was generated by ChatGPT to emphasize that this is a personal reflection, not a formal scientific critique).

That format clearly marks my contribution while keeping your authorship front and center. It’ll look neat and professional on your site.

You said:

Instead of my, could you use Maria’s

ChatGPT said:

Absolutely! Here’s the revised version with your requested change:

ChatGPT-assisted preface to the blog article

This post reflects Maria’s personal views and experiences as an observational astronomer navigating the era of computational astronomy (the accompanying commentary was generated by ChatGPT to emphasize that this is a personal reflection, not a formal scientific critique).

It clearly attributes the reflections to you while still marking ChatGPT’s contribution.

You said: